Frequency modulation method and device

CN118974715APending Publication Date: 2024-11-15HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202280094627.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-08-24
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In the chip system, when the SoC domain clock frequency is adjusted, the clock frequency switching delay of the interconnection interface is large, resulting in an increase in service delay and an increase in the cache space of service data, making it difficult to achieve fast frequency switching.

Method used

By exchanging frequency-level information between dies, the circuit configuration parameters of each frequency level are trained in advance to achieve rapid switching of frequency gears between dies, reduce link switching delays, and dynamically adjust reception when the SoC domain frequency changes. The operating frequency of the circuit.

Benefits of technology

It achieves fast switching of SoC domain and interconnection interface frequencies, reduces the cache space and delay of business data, reduces interface power consumption, and dynamically adjusts the voltage during frequency switching to reduce power consumption.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118974715A_ABST
    Figure CN118974715A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a frequency modulation method and device, relates to the technical field of chips, and can realize fast switching of clock frequencies of an SoC (System on Chip) domain and an interconnection interface in a Chiplet system and reduce a cache space required by service data. The method is applied to a chip system, the chip system comprises a first bare chip and a second bare chip, and the method comprises the steps that the first bare chip sends a frequency switching request to the second bare chip, and the frequency switching request comprises information indicating a switched frequency level; and the second bare chip adjusts the working frequency of a receiving circuit of the second bare chip according to the frequency level. The embodiment of the invention is used for carrying out frequency modulation on the bare chip in the chip system.
Need to check novelty before this filing date? Find Prior Art

Description

Frequency modulation method and device Technical Field

[0001] The present application relates to the field of chip technology, and in particular to a frequency modulation method and device. Background Art

[0002] In a chip system (chiplet), to achieve low latency across the entire chiplet, the clock frequency of the die-to-die interconnect interface within the chiplet can be synchronized with the clock frequency of the SoC domain to reduce the latency caused by asynchronous processing. In this case, when the SoC domain clock frequency is adjusted, the clock frequency of the interconnect interface also needs to be adjusted synchronously.

[0003] During dynamic frequency modulation, service data still exists. When the SoC domain clock frequency switches, service data needs to be cached in the system. Long switching delays not only increase service latency but also increase the cache space for service data. Therefore, reducing the clock frequency switching delay in chiplets is an urgent problem to be solved.

[0004] Summary of the Invention

[0005] The embodiments of the present application provide a frequency modulation method and apparatus that can achieve rapid switching of the clock frequencies of the SoC domain and the interconnection interface in a chiplet system, thereby reducing the cache space required for service data.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions:

[0007] In a first aspect, a frequency modulation method is provided. The method is applied to a chip system, the chip system including a first bare chip and a second bare chip. The method includes: the first bare chip sends a frequency switching request to the second bare chip, the frequency switching request including information indicating the frequency level after switching; the second bare chip adjusts the operating frequency of the receiving circuit of the second bare chip according to the frequency level.

[0008] The frequency levels in this application may also be referred to as frequency gears. This application is equivalent to pre-training the frequency levels between the bare chips in the chip system, with each frequency level corresponding to an operating frequency. When a first bare chip and a second bare chip are transmitting service data, if the operating frequency of the transmitting circuit of the first bare chip switches, the second bare chip needs to adaptively adjust the operating frequency of the receiving circuit. When the first bare chip sends information indicating the switched frequency level to the second bare chip, the second bare chip can quickly adjust the operating frequency of the receiving circuit corresponding to the frequency level based on the information about the switched frequency level.

[0009] Therefore, compared to the problem of long switching delays caused by the need to enter the recovery state for link retraining each time when switching the rate in the PCIE link and then switching back to the L0 state, the present application has a shorter delay when performing frequency switching. In addition, compared to the problem of the inability to optimize interface power consumption caused by adjusting the number of lanes between interconnected interfaces to switch the rate between bare chips, the present application directly implements operating frequency switching by exchanging information indicating the frequency level between bare chips. When performing operating frequency switching, the interface power consumption can be adjusted accordingly.

[0010] Moreover, usually, when performing frequency switching between bare chips, in order to keep the service flow uninterrupted, the service data needs to be cached. Based on the realization of fast frequency switching between bare chips, the present application can also reduce the cache space required for service data caching.

[0011] In one possible design, the transmit circuitry of the first die and the receive circuitry of the second die are in the same clock domain. Alternatively, the SoC domain, transmit circuitry, and receive circuitry of the first die are all in the same clock domain. This allows the digital logic on the data link layer of the first and second dies to dynamically adjust voltage based on the SoC domain's frequency change when the operating frequency of the transmit circuitry is modulated, further reducing power consumption.

[0012] In one possible design, before the first bare chip sends a frequency switching request to the second bare chip, the method also includes: the first bare chip receives a frequency mode configuration command sent by the application layer, the frequency mode configuration command includes first indication information, and the first indication information is used to indicate the frequency level after switching; the first bare chip generates a frequency switching request according to the frequency mode configuration command.

[0013] In this design, when the service type of the first die's application layer changes, the operating frequency of the service flow processed by the first die for that service type will also change accordingly, adapting the changed operating frequency to the new service type. Therefore, when the first die's application layer determines that the first die's operating frequency needs to be changed, it can send a frequency mode configuration command to the first die, indicating the frequency level to be switched. For example, the application layer stores the frequency levels corresponding to different service types.

[0014] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: the first die stops receiving service data from the application layer.

[0015] In other words, when the application layer of the first die determines to perform frequency modulation, the application layer of the first die stops sending service data to the SoC domain of the first die so that the first die can adjust the frequency and the frequency level after interactive switching with the second die.

[0016] In one possible design, the first die sending a frequency switching request to the second die includes: the first die sending the frequency switching request to the second die through a first low power control word LPW flow control unit or a first sideband control command; the method also includes: the first die receiving a frequency switching response sent by the second die through a second LPW flow control unit or a second sideband control command.

[0017] The first LPW flit can be understood as a command sent from the first die to the second die via a data link. The first sideband control command can be understood as a command sent from the first die to the second die via a sideband control channel. Correspondingly, the second LPW flit can be understood as a command sent from the second die to the first die via a data link. The second sideband control command can be understood as a command sent from the second die to the first die via a sideband control channel.

[0018] In one possible design, the first LPW flow control unit includes a flow control unit header field, a flow control unit content field, and a flow control unit tail field; wherein the flow control unit content field includes a type subfield and a command subfield, the type subfield is used to indicate that the first LPW flow control unit is a frequency switching command, and the command subfield is used to indicate the frequency level.

[0019] In one possible design, the first sideband control command includes a field indicating a message type, a field indicating a message encoding, a field indicating a path identifier, and a field indicating a check bit; wherein the field indicating the message type is used to indicate that the first sideband control command is a frequency switching command.

[0020] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: the first die stops sending the associated clock to the second die, and adjusts the operating frequency of the transmitting circuit of the first die according to the frequency level.

[0021] In this design, a clock lane can exist between the first die and the second die. The receiving circuit in the second die receives service data sent by the transmitting circuit of the first die at the clock frequency of the associated clock sent by the first die via this clock lane. If the first die needs to switch frequencies, it can stop sending the associated clock to the second die. After the operating frequency of the transmitting circuit of the first die is switched, the associated clock at the new operating frequency can be sent to the second die via this clock lane at the new operating frequency. This adjusts the clock frequency of the receiving circuit of the second die accordingly.

[0022] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: the first die continuously sending an empty flow control unit to the second die until receiving a frequency switching response sent by the second die.

[0023] Alternatively, the first die continues to send frequency switching commands to the second die until receiving a frequency switching response sent by the second die.

[0024] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: after the first die determines that the frequency switching negotiation with the second die is completed, responding to the application layer that the frequency switching negotiation is completed.

[0025] That is, after the first bare chip receives the frequency switching configuration command sent by the application layer, if the first bare chip completes the frequency switching negotiation with the second bare chip, it can respond to the application layer of the first bare chip that the frequency switching negotiation is completed, so that the application layer can start the frequency switching of the SoC domain of the first bare chip.

[0026] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: the first die sending a frequency switching instruction to the second die, where the frequency switching instruction is used to instruct the second die to perform frequency switching.

[0027] When the first die sends a frequency switching instruction to the second die, the first die may initiate, be in the process of, or complete the frequency switching of its transmit circuit. When the second die receives the frequency switching instruction, it may initiate the frequency switching of its receive circuit.

[0028] In this application, the first die performing frequency switching of the first die's transmit circuit can be understood as the first die determining circuit configuration parameters of the transmit circuit based on the frequency level and configuring the transmit circuit parameters based on the transmit circuit configuration parameters. Similarly, the second die performing frequency switching of the second die's receive circuit can be understood as the second die determining circuit configuration parameters of the receive circuit based on the frequency level and configuring the receive circuit parameters based on the receive circuit configuration parameters.

[0029] In one possible design, the method further includes: after the first die waits for a fixed delay, starting a clock associated with the first die.

[0030] Typically, this fixed delay should cover the stabilization time of the second die after switching its operating frequency, as well as the stabilization time of the first die after switching its operating frequency. This gives the first and second dies enough time to complete the frequency switch. After both die complete the frequency switch, the first die can start the associated clock at the new operating frequency and begin exchanging service data with the second die.

[0031] In one possible design, the method further includes: the first die sending data to the second die at the frequency level; and the second die receiving the data sent by the first die at the frequency level.

[0032] In this way, after the first bare chip and the second bare chip complete the switching of the operating frequency through the exchange of frequency-level information, they exchange service data at the new operating frequency.

[0033] In a second aspect, a frequency modulation method is provided, which is applied to a chip system, the chip system including a first bare chip and a second bare chip, the method including: the second bare chip receives a frequency switching request sent by the first bare chip, the frequency switching request including information indicating the frequency level after switching; the second bare chip adjusts the operating frequency of the receiving circuit of the second bare chip according to the frequency level.

[0034] The beneficial effects of the second aspect can be found in the description of the first aspect.

[0035] In one possible design, the transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

[0036] In one possible design, the second die receives the frequency switching request sent by the first die, including: the second die receives the frequency switching request sent by the first die through a first low power control word LPW flow control unit or a first sideband control command; the method also includes: the second die sends a frequency switching response to the first die through a second LPW flow control unit or a second sideband control command.

[0037] In one possible design, the method further includes: when the second bare chip receives the frequency switching request sent by the first bare chip, recording information indicating the frequency level after switching.

[0038] In one possible design, before the second die adjusts the operating frequency of the receiving circuit of the second die according to the frequency level, the method further includes: the second die receiving a frequency switching instruction sent by the first die, where the frequency switching instruction is used to instruct the second die to perform frequency switching.

[0039] In a third aspect, a frequency modulation method is provided, which is applied to a chip system, the chip system including a first bare chip and a second bare chip, the method including: the first bare chip sends a frequency switching request to the second bare chip, the frequency switching request including information indicating the frequency level after switching; the first bare chip adjusts the operating frequency of the transmitting circuit of the first bare chip according to the frequency level.

[0040] The beneficial effects of the third aspect can be found in the description of the first aspect.

[0041] In one possible design, the transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

[0042] In one possible design, the first die sending a frequency switching request to the second die includes: the first die sending the frequency switching request to the second die through a first low power control word LPW flow control unit or a first sideband control command; the method also includes: the first die receiving a frequency switching response sent by the second die through a second LPW flow control unit or a second sideband control command.

[0043] In a fourth aspect, a chip system is provided, which includes a first bare chip and a second bare chip, wherein: the first bare chip is used to send a frequency switching request to the second bare chip, and the frequency switching request includes information indicating the frequency level after switching; the second bare chip is used to adjust the operating frequency of the receiving circuit of the second bare chip according to the frequency level.

[0044] The beneficial effects of the fourth aspect can be found in the description of the first aspect.

[0045] In one possible design, the transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

[0046] In one possible design, the first bare chip is used to: receive a frequency mode configuration command sent by the application layer, the frequency mode configuration command includes first indication information, and the first indication information is used to indicate the frequency level after switching; and generate a frequency switching request according to the frequency mode configuration command.

[0047] In one possible design, the first bare chip is configured to: send a frequency switching request to the second bare chip through a first low power control word LPW flow control unit or a first sideband control command; and receive a frequency switching response sent by the second bare chip through a second LPW flow control unit or a second sideband control command.

[0048] In a fifth aspect, a second bare chip is provided, the chip system includes a first bare chip and a second bare chip, the second bare chip includes: a receiver for receiving a frequency switching request sent by the first bare chip, the frequency switching request including information indicating the frequency level after switching; a processor for adjusting the operating frequency of the receiving circuit of the second bare chip according to the frequency level.

[0049] In one possible design, the transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

[0050] In one possible design, the receiver is configured to: receive a frequency switching request sent by the first bare chip through a first low power control word LPW flow control unit or a first sideband control command; and the receiver is further configured to: send a frequency switching response to the first bare chip through a second LPW flow control unit or a second sideband control command.

[0051] In one possible design, the processor is further configured to: when the receiver is configured to receive a frequency switching request sent by the first bare chip, record information indicating a frequency level after switching.

[0052] In one possible design, the receiver is further used to: receive a frequency switching indication sent by the first bare chip, where the frequency switching indication is used to instruct the second bare chip to perform frequency switching.

[0053] In a sixth aspect, a first bare chip is provided, the chip system includes the first bare chip and a second bare chip, the first bare chip includes: a transmitter for sending a frequency switching request to the second bare chip, the frequency switching request including information indicating the frequency level after switching; a processor for adjusting the operating frequency of the transmitting circuit of the first bare chip according to the frequency level.

[0054] In one possible design, the transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

[0055] In one possible design, the transmitter is configured to send a frequency switching request to the second bare chip via a first low power control word LPW flow control unit or a first sideband control command; and further includes a receiver configured to receive a frequency switching response sent by the second bare chip via a second LPW flow control unit or a second sideband control command.

[0056] In the seventh aspect, an embodiment of the present application provides a chip system, which includes a bare chip as described in any possible design of the first to third aspects; the chip system includes multiple bare chips, each bare chip includes an interface circuit and a processor (SoC domain); the interface circuit and the processor are interconnected through lines; the processor receives and executes computer instructions from a memory outside the chip system through the interface circuit.

[0057] In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium comprising computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the method described in the first aspect and any possible design of the first aspect.

[0058] In a ninth aspect, an embodiment of the present application provides a computer-readable storage medium comprising computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the method described in the second aspect and any possible design of the second aspect.

[0059] In the tenth aspect, an embodiment of the present application provides a computer-readable storage medium, including computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the method described in the third aspect and any possible design of the third aspect.

[0060] In the eleventh aspect, an embodiment of the present application provides a computer program product, which, when running on a computer or a processor, enables the computer or the processor to execute the method in the above-mentioned first aspect and any possible implementation manner.

[0061] In a twelfth aspect, an embodiment of the present application provides a computer program product. When the computer program product runs on a computer or a processor, it enables the computer or the processor to execute the method in the above-mentioned second aspect and any possible implementation method.

[0062] In a thirteenth aspect, an embodiment of the present application provides a computer program product, which, when executed on a computer or a processor, enables the computer or the processor to execute the method in the third aspect and any possible implementation thereof.

[0063] It can be understood that any of the first bare chip, second bare chip, chip system, computer-readable storage medium or computer program product provided above can be applied to the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method and will not be repeated here.

[0064] These and other aspects of the present application will become more readily apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] FIG1 is a schematic diagram of a circuit for inter-die coupling provided by an embodiment of the present application;

[0066] FIG2 is a schematic diagram of a state switching of frequency switching in a PCIE link provided by an embodiment of the present application;

[0067] FIG3 is a schematic diagram of achieving a switching rate between dies by adapting different lanes according to an embodiment of the present application;

[0068] FIG4 is a schematic diagram of a chip system provided in an embodiment of the present application;

[0069] FIG5 is a flow chart of a frequency modulation method provided in an embodiment of the present application;

[0070] FIG6 is a circuit diagram of an interconnection interface between bare chips and a SoC domain coupling according to an embodiment of the present application;

[0071] FIG7 is a flow chart of a frequency modulation method provided in an embodiment of the present application;

[0072] FIG8 is a schematic diagram of a sideband signal interface interconnection between dies provided by an embodiment of the present application;

[0073] FIG9 is a timing diagram of a sideband signal interface provided in an embodiment of the present application when transmitting in a message manner;

[0074] FIG10 is a schematic diagram of a format of a sideband control signal provided in an embodiment of the present application;

[0075] FIG11 is a schematic structural diagram of a bare chip provided in an embodiment of the present application;

[0076] FIG12 is a schematic structural diagram of a bare chip provided in an embodiment of the present application;

[0077] FIG13 is a schematic diagram of the structure of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0078] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships can exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0079] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this embodiment, unless otherwise specified, "plurality" means two or more.

[0080] A chiplet system, also known as a chip system, is a combination of dies with different functions, and in a sense, different IPs. Starting from the system end, complex functions are first decomposed, and then a variety of dies with single specific functions are developed, which can be modularly assembled with each other, such as data storage, computing, signal processing, and data flow management. Ultimately, a chiplet chip network is established based on this. Alternatively, it can be understood as a model in which multiple module chips are packaged together with the underlying basic chip through die-to-die internal interconnect technology to form a multifunctional heterogeneous System in Packages (SiPs) chip model.

[0081] In chiplet systems, low latency and low power consumption are key features. In actual work scenarios, the bandwidth of the die-to-die interconnect interface varies with the service bandwidth. Interface power consumption also varies with bandwidth, achieving proportional power reduction at low service bandwidths.

[0082] At the same time, to achieve low latency across the entire die-to-die path, the clock frequency of the interconnect interface can be synchronized with the SoC domain within the die to minimize latency increases due to asynchronous processing. As shown in Figure 1, the die includes the SoC domain and the interconnect interface circuit. The interconnect interface circuit includes a transmitting circuit, a receiving circuit, and a sideband control channel. The SoC domain can transmit application data (i.e., service data) to the transmitting circuit of the interconnect interface circuit. The SoC domain can also transmit control signals via the sideband control channel of the interconnect interface circuit.

[0083] The SoC domain in die 1, the transmitting circuitry of die 1 in the interconnect interface, and the receiving circuitry of die 2 in the interconnect interface are in the same clock domain. When the clock frequency of the SoC domain in die 1 changes, the clock frequencies of the transmitting circuitry in the interconnect interface of die 1 and the receiving circuitry in the interconnect interface of die 2 also need to be adjusted synchronously. The transmitting circuitry in the interconnect interface of die 1 can serve as the link master for die 1, and the receiving circuitry in the interconnect interface of die 2 can serve as the link slave for die 2.

[0084] Similarly, the SoC domain in die 2, the transmit circuitry in die 2's interconnect interface, and the receive circuitry in die 1's interconnect interface are in the same clock domain. When the clock frequency of the transmit circuitry in die 2's SoC domain changes, the clock frequencies of the transmit circuitry in die 2's interconnect interface and the receive circuitry in die 1's interconnect interface also need to be adjusted synchronously. The transmit circuitry in die 2's interconnect interface can function as the link master for die 2, and the receive circuitry in die 1's interconnect interface can function as the link slave for die 1.

[0085] Figure 1 also shows that the SoC domain in die 1 and the receiving circuit in the interconnect interface of die 1 are coupled via an asynchronous circuit. Considering that asynchronous circuits are primarily combinational logic circuits and do not use clock pulses for synchronization, a handshake protocol is generally used to achieve data synchronization, communication, and operation order between different components. When the asynchronous circuit receives data from the receiving circuit, it must send the data to the SoC domain in die 1 according to the handshake protocol. Therefore, the clocks of the SoC domain in die 1 and the receiving circuit in die 1 are in an asynchronous clock domain. Similarly, the clocks of the SoC domain in die 2 and the receiving circuit in the interconnect interface of die 2 are also in an asynchronous clock domain. Die 1 and die 2 can transmit sideband signals via a sideband control channel. When die 1 is the master chip and die 2 is the slave chip, the transmitting circuit in die 1 can send downstream data to the receiving circuit in die 2 via a downstream link, and the transmitting circuit in die 2 can send upstream data to the link slave in die 1 via an upstream link. Die 1 and die 2 can transmit upstream data and downstream data via a flow control unit (flit).

[0086] In addition, during the frequency modulation process, service data still exists and needs to be cached during the switching process. Therefore, reducing the switching delay helps to reduce the cache space and also reduces the service delay caused by the switching process.

[0087] Currently, in high-speed interconnect solutions, to maintain latency stability and link state transmission performance after switching between different clock frequencies, in high-speed peripheral component interconnect express (PCIE) links, when the clock frequency changes, the state machine needs to switch from the L0 state to the recovery state for link retraining. After training is complete, it switches back to the L0 state. As shown in Figure 2, the L0 state is the normal operating state of the PCIE link, in which the PCIE link can send and receive data normally.

[0088] However, this solution requires a long retraining process, which results in a longer time for link stabilization when switching between different clock frequencies. Furthermore, service data persists during the frequency switching process, increasing the cache space required for service data. This solution is not suitable for scenarios where real-time link switching is critical.

[0089] In a scheme for matching rate switching, the purpose of switching rate can be achieved by adapting the number of different paths (lanes). However, in this scenario, as shown in Figure 3, the interconnect interface and the SoC domain are coupled through an asynchronous circuit. The clock domain of the interconnect interface in a single bare chip is different from the clock domain of the SoC domain, that is, an asynchronous clock domain. The clock domain of the interconnect interface and the clock domain of the SoC domain do not use clock pulse synchronization. The interconnect interface operates at a fixed frequency. When the clock frequency of the SoC domain is adjusted, the clock frequency of the interconnect interface remains unchanged, and the rate of the interconnect interface is matched by adjusting the number of lanes of the interconnect interface. In this scheme, this asynchronous processing will increase the delay. In addition, the interconnect interface matches the change in rate according to the number of lanes. When the rate of the interconnect interface decreases, the digital part of the interconnect interface cannot follow the rate reduction to reduce the operating voltage, and the power consumption performance of the interconnect interface is poor.

[0090] Therefore, this application provides a frequency modulation method that can dynamically adjust the interconnection interface frequency in a chiplet. By exchanging different frequency level indication information between bare chips, it can achieve rapid switching of frequency levels and reduce link switching latency.

[0091] In the frequency modulation method of the present application, when the operating frequency of the SoC domain changes, the operating frequency of the interconnection interface between the bare chips also changes accordingly. Before frequency modulation, the circuit configuration parameters of the interconnection interface at each frequency level can be trained separately during the initialization process, and the training results are saved in each bare chip. When the transmitting end needs to adjust the frequency of the interconnection interface according to the operating frequency of the SoC domain, the transmitting end informs the receiving end of the indication of the frequency level to be switched through the interactive process. The receiving end can query the corresponding circuit configuration parameters from the training parameters according to the indication of the frequency level to switch to the frequency level.

[0092] The embodiments of the present application can be applied to a chiplet system. In the chiplet system, a plurality of bare chips with specific functions that can be modularly assembled with each other are included. For example, the bare chip can realize functions such as data storage, computing, signal processing, and data flow management. Based on the rich IP reserves, the chiplet system proposes the concept of IP chipization (IP as a Chiplet, IaaC), which aims to realize the "plug and play" of special function IP with bare chips. For example, for some core processor IPs, the bare chip can be a graphics processing unit (GPU) IP, a neural network processor (NPU) IP, a video processor (VPU) IP, a digital signal processor (DSP) IP, an image signal processor (ISP) IP, and a display processor IP, etc. In addition, the bare chip can also be a mixed digital-analog IP and a radio frequency IP, etc.

[0093] The chiplet system in this application, i.e., the chip system, can also be referred to as a chiplet system or a core particle system. In a chiplet system, a bare die can also be referred to as a chiplet, and in a core particle system, a bare die can also be referred to as a core particle.

[0094] For example, the chiplet system can be applied to a system-on-a-chip (SoC), where multiple functional dies are integrated on the SoC. For example, the chip system includes a first die and a second die as shown in FIG4 . The first die can transmit service data / application data to the second die via a data channel, where the data channel includes the downstream link and upstream link described above. The first die can transmit control signals to the second die via a sideband control channel.

[0095] When the operating frequency of the transmitting circuit of the first die changes, information indicating the switched frequency level can be sent to the second die. The second die can adjust the operating frequency of the receiving circuit according to the frequency level, achieving fast frequency switching between the die.

[0096] In some embodiments, the first die and the second die can be understood as two chips with different functions in initialization and power management. For example, the first die is a host processor chip. The host can initiate a flow to enter a low-power state and perform frequency switching in conjunction with the low-power state. The second die is a chip connected to the host processor chip, acting as a slave processor chip. For example, the second die is a memory chip. A slave processor chip generally cannot initiate a flow to enter a low-power state, but can respond to the host processor chip's request for a flow to enter a low-power state.

[0097] The frequency modulation method of this application is introduced below.

[0098] An embodiment of the present application provides a frequency modulation method, which is applied to a chip system (chiplet system). The chip system includes a first die and a second die. As shown in FIG5 , the method includes the following process.

[0099] 501. A first bare chip sends a frequency switching request to a second bare chip, where the frequency switching request includes information indicating a frequency level after switching.

[0100] For example, when the operating frequency of the SoC domain in the first die is switched, information indicating the switched frequency level may be sent to the second die so that the second die also synchronously switches the operating frequency of the receiving circuit.

[0101] The frequency level can also be understood as the frequency gear.

[0102] In some embodiments, the frequency level in this application indicates a preset frequency mode or frequency parameter corresponding to a service. The second die may pre-store circuit configuration parameters corresponding to the frequency mode or frequency parameter. Thus, when the second die determines the corresponding circuit configuration parameters based on the frequency mode or frequency parameter, it can adjust device parameters in the receiving circuit based on the determined circuit configuration parameters, thereby adjusting the operating frequency of the receiving circuit so that the receiving circuit can process data at the corresponding service rate.

[0103] Exemplarily, the device parameters here include parameters such as capacitance, resistance, inductance or register configuration, and may also include parameters related to the receiving filter and the receiving level.

[0104] 502. The second die adjusts the operating frequency of the receiving circuit of the second die according to the frequency level.

[0105] The second bare chip adjusts the operating frequency of the receiving circuit according to the frequency level, which can be understood as the second bare chip adjusting the circuit configuration parameters of the receiving circuit according to the frequency level.

[0106] In some embodiments, multiple training parameters may be obtained through pre-training, each training parameter including circuit configuration parameters corresponding to a frequency level. When the second die receives information indicating the frequency level after switching, it may determine the corresponding circuit configuration parameters based on the information indicating the frequency level after switching, and complete circuit configuration based on the circuit configuration parameters.

[0107] After the transmission circuit of the first die and the receiving circuit of the second die have completed the adjustment of the circuit configuration parameters, the first die and the second die switch frequencies. The transmission circuit of the first die sends data at the switched frequency, and the receiving circuit of the second die receives data at the switched frequency.

[0108] Therefore, in this application, when the frequency levels between bare chips in the chiplet system and the parameters for adjusting the operating frequency are trained in advance, the frequency level switching information is exchanged between bare chips. When the bare chips switch the operating frequency, the stored training parameters can be called according to the information indicating the frequency level after switching to complete the rapid frequency switching.

[0109] In some embodiments, in the frequency modulation method of the present application, the interconnection interface between the bare chips and the SoC domain are in the same clock domain, and the SoC domain does not need to perform asynchronous processing when transmitting data through the interconnection interface. As shown in Figure 6, the SoC domain of the first bare chip, the transmitting circuit in the first bare chip, and the receiving circuit in the second bare chip are in the same clock domain, which is recorded as clock domain 1. When the SoC domain in the first bare chip performs clock frequency switching, the transmitting circuit in the first bare chip and the receiving circuit in the second bare chip need to perform clock frequency switching. The first bare chip can configure the transmitting circuit of the first bare chip according to the pre-stored training parameters, that is, the circuit configuration parameters corresponding to the frequency level, and the second bare chip also configures the receiving circuit of the second bare chip according to the circuit configuration parameters corresponding to the frequency level.

[0110] Similarly, the SoC domain in the second die, the transmitter circuitry in the second die, and the receiver circuitry in the first die are in the same clock domain, designated as Clock Domain 2. When the SoC domain in the second die switches clock frequencies, both the transmitter circuitry in the second die and the receiver circuitry in the first die need to switch clock frequencies. The first die can configure the transmitter circuitry of the first die based on pre-stored training parameters, specifically circuit configuration parameters corresponding to the frequency level, and the receiver circuitry of the second die can be configured based on the circuit configuration parameters corresponding to the frequency level.

[0111] In some embodiments, when the first bare chip determines according to the application layer that frequency switching is required, the transmitting circuit in the first bare chip, the receiving circuit in the first bare chip, the receiving circuit in the second bare chip, and the transmitting circuit in the second bare chip may all be reconfigured in terms of operating frequency, that is, the circuit configuration parameters may all be reconfigured.

[0112] The following further describes the embodiments of the present application.

[0113] An embodiment of the present application provides a frequency modulation method, which is applied to a chip system (chiplet system). The chip system includes a first die and a second die. As shown in FIG7 , the method includes the following process.

[0114] 701. A first bare chip receives a frequency mode configuration command sent by an application layer, where the frequency mode configuration command includes first indication information, and the first indication information is used to indicate a frequency level after switching.

[0115] The application layer can be understood as the application service of the first bare chip.

[0116] For example, when the application layer determines that the service type has changed and that service data of the service type needs to be transmitted at a new frequency level, the application layer may send a frequency mode configuration command to the first die, indicating the switched frequency level to the first die. For example, the application layer sends the frequency mode configuration command to the SoC domain of the first die.

[0117] 702. The first die generates a frequency switching request according to the frequency mode configuration command, and sends the frequency switching request to the second die via a first low power word (LPW) flit or a first sideband command, where the frequency switching request includes information indicating a frequency level after switching.

[0118] Correspondingly, the second die receives the frequency switching request sent by the first die through the first LPW flit or the first sideband control command.

[0119] Among them, LPW can be understood as a signal that communicates through low power consumption, low data rate and low cost.

[0120] A sideband refers to the modulated signal. It's a frequency band generated above and below the center carrier frequency, called a sideband. Sideband control commands are control commands transmitted via the sideband. These commands are typically transmitted using a sideband address.

[0121] In some embodiments, the first die stops sending service data to the second die when sending the first LPW flit. After sending the frequency switching request, the first die may continue to send service data to the second die normally.

[0122] In some embodiments, when the first die sends the first sideband control command, it can normally send service data to the second die, that is, the normal sending of service data is not affected.

[0123] In some embodiments, the SoC domain of the first die sends a first LPW flit to the receiving circuit of the second die via the transmitting circuit of the first die. The first LPW flit is equivalent to a handover request in this application. With reference to FIG6 , the first LPW flit is sent by the transmitting circuit of the first die to the second die via the downstream link.

[0124] The downstream link and the upstream link may be high-speed data channels. For example, when the interconnection interface is a high-speed serial interface, such as PCIE or Universal Serial Bus (USB), PCIE or USB can transmit service data between dies via the high-speed data channel.

[0125] In this case, taking the high-speed serial interface as PCIE as an example, the PCIE of the first bare chip can transmit the first LPW flit to the PCIE of the second bare chip through the high-speed data channel.

[0126] Illustratively, the first LPW flit includes a flit header field, a flit payload field, and a flit tail field. The flit payload field includes a type subfield (SUB_TYPE) and a command subfield (CMD). The type subfield is used to indicate that the first LPW flit is a frequency switching command, and the command subfield is used to indicate the frequency level. Table 1 shows an example format of an LPW flit.

[0127] Table 1 LPW flit format

[0128]

[0129] When the bit value of the Type subfield in the Flit Header field is "0", it indicates that the first LPW flit is a control flit, which can be used to instruct the second bare chip to execute the control command indicated by the first LPW flit. When the bit value of the SUB_TYPE subfield in the Flit payload field is "1100", it indicates that the first LPW flit is a frequency switching command. The bit value of the CMD subfield in the Flit payload field can also be one of multiple bit values ​​('b00000000 to 'b11111111), used to indicate the frequency level N, where N is an integer. For example, when the bit value of N in the CMD subfield is 00000001, the value of N is 1, and the operating frequency indicated by frequency level 1 is 2 / 256 of the highest frequency.

[0130] This application does not limit the bit value of the CMD subfield used to indicate the frequency level.

[0131] In some embodiments, when the first die sends a frequency switching request to the second die through the first sideband control command, it can be understood that the SoC domain in the first die sends the frequency switching request to the sideband control channel of the second die through the sideband control channel of the first die.

[0132] Exemplarily, the first sideband control command can be exchanged via a sideband signal interface between dies. For example, for a single die, the sideband signal interface includes SBC_TXC for transmitting the clock frequency of the transmitting side, SBC_TXD for transmitting the serial data of the transmitting side, SBC_RXC for transmitting the clock frequency of the receiving side, and SBC_RXD for transmitting the serial data of the receiving side.

[0133] In summary, the sideband signal interface is used for the interaction of control signals and status between two dies. The control signal is, for example, the first sideband control command in this application. In a chiplet system, the sideband signal interface can use a serial transmission method. The sideband signal of each die includes a sideband signal in the transmitting direction and a sideband signal in the receiving direction. Each sideband signal can be connected using one clock signal and one data signal.

[0134] Figure 8 shows a schematic diagram of the sideband signal interface interconnection between dies. The description of each sideband signal interface can be found in Table 2.

[0135] Table 2 Sideband signal interface

[0136]

[0137] In Table 2, SBC_TXC represents the clock signal of the sideband control signal in the transmitting direction; SBC_TXD represents the data (control command) of the sideband control signal in the transmitting direction; SBC_RXC represents the clock signal of the sideband control signal in the receiving direction; and SBC_RXD represents the data (control command) of the sideband control signal in the receiving direction.

[0138] In the example of the sideband signal interface shown in Table 2, the first die can send the first sideband control signal to the SBC_RXD of the second die through SBC_TXD. The first die sends the first sideband control signal at the clock frequency indicated by SBC_TXC, and the second die receives the first sideband control signal at the clock frequency indicated by SBC_RXC.

[0139] The clock frequencies SBC_TXC and SBC_RXC of the sideband signals may be in a synchronous timing relationship with the sideband control signal (data), with the rising edge of the clock and the transition edge of the data being aligned.

[0140] Figure 9 shows a timing diagram of a sideband signal interface transmitting in message form. As shown in Figure 9, one bit of data can be transmitted using a single data rate (SDR) per clock cycle. Each continuous transmission of 16 bits of data (D0-D15) (16 clock cycles) constitutes a packet. Two packets (for example, packets 0 and 1 in Figure 9) are separated by at least eight clock cycles. During the interval between packets, SBC_TXC can be maintained at a low level, and the data I / O (SBC_TXD) can be in a low-power state.

[0141] The first sideband control command in the present application can be sent to the second die via these 16 clock cycles, occupying 16 bits.

[0142] For example, the message format of the sideband signal of the present application, for example, when the format of the first sideband control signal is 16 bits, can refer to FIG10. The function of each field in this format can be seen in the description of Table 3.

[0143] Table 3

[0144]

[0145] Applying the examples of FIG10 and Table 3, when the first sideband control command of the present application is transmitted in a 16-bit message format, the bit value of the type[2:0] field may be 100.

[0146] For example, the definition of the operation code (opcode) of the frequency switching command of the data link layer, ie, type=100, can be found in Table 4, and the implementation method is similar to that of the CMD field in Table 1.

[0147] Table 4

[0148]

[0149] In some embodiments, the definition of the opcode of other control information of the data link layer in Table 3, that is, when type=111, can be found in Table 5.

[0150] Table 5

[0151] opcode[7:0] value function description 11111111SBC_SYNC_IND: data synchronization indication. Others are reserved

[0152] The SBC_SYNC_IND signal may be used, for example, by the first die to notify the second die that the frequency switching of the first die is complete, and the second die may start the frequency switching synchronously.

[0153] 703. When receiving the frequency switching request, the second bare chip records information indicating the frequency level after switching.

[0154] In some embodiments, corresponding to step 702, if the second die receives the first LPW flit, the receiving circuit in the second die stops receiving service data from the downstream link and records information indicating the frequency level after the frequency switch. After the second die stops receiving service data, if the second die completes the frequency switch, the second die can receive service data of the service type at the new operating frequency.

[0155] Corresponding to step 702 , if the second die receives the first sideband control command, the second die may receive service data normally and record information indicating the frequency level after switching.

[0156] 704. The second die sends a frequency switching response to the first die.

[0157] Exemplarily, the second die sends a frequency switching response to the first die via a second LPW flit or a second sideband control command.

[0158] Accordingly, the first die receives the frequency switching response sent by the second die through the second LPW flit or the second sideband control command.

[0159] At this point, the first die may be waiting to receive a frequency switch response.

[0160] Exemplarily, the format of the second LPW flit is similar to the format of the first LPW flit shown in Table 1, except that, referring to Table 1, the bit value of SUB_TYPE is 4'b1111, indicating that the second LPW flit is a response command for switching frequency.

[0161] The format of the second sideband control command is similar to the format of the sideband signal shown in Table 3, wherein the bit value of the type[2:0] field can be 011.

[0162] In some embodiments, the second die may stop sending service data to the first die while sending the second LPW flit, and may continue sending service data to the first die normally after completing sending the second LPW flit.

[0163] When the second bare chip sends the second sideband control command, it can normally send service data to the first bare chip, that is, the normal service data sending of the second bare chip is not affected.

[0164] In some embodiments, when the first die issues a frequency switching request, a timer may be started. If the first die does not receive a frequency switching response within the timer, the first die may resend the frequency switching request via the data link (downstream link). Alternatively, the first die may continue to send empty flow control units (NULL flit) to the second die via the data link until a frequency switching response is received from the second die.

[0165] The timing time is greater than the two-way transmission delay of the interconnection interface and can be configured during the specific implementation process.

[0166] When the first die continues to send frequency switching requests or NULL flits within a predetermined time, the services between the first die and the second die can be kept uninterrupted, and the first die and the second die can cache the received service data.

[0167] In some embodiments, when the second die receives the first LPW flit, the second die may adjust the priority of the second LPW flit to be sent to the highest priority, so as to preferentially send the second LPW flit to the first die.

[0168] 705 . After the first bare chip determines that the frequency switching negotiation with the second bare chip is completed, the first bare chip responds to the application layer that the frequency switching negotiation is completed.

[0169] The above steps 701 to 704 can be understood as completing the frequency switching negotiation at the data link layer between the first die and the second die.

[0170] When the first die receives the frequency switching response, it determines that the frequency negotiation with the second die is complete and can respond to the application layer of the first die that the frequency switching negotiation is complete. The response here can be understood as a response to the frequency mode configuration command in step 701.

[0171] 706 . The application layer of the first bare chip stops sending service data to the first bare chip, and switches the operating clock frequency of the SoC domain of the first bare chip.

[0172] Accordingly, the data link layer of the first die stops receiving service data from the application layer. For example, the first die may cache service data that has been received but not yet sent to the second die. When the frequency switch is complete, the SoC domain of the first die resumes sending service data (service packets) to the second die.

[0173] The operating clock frequency of the SoC domain of the first die switched here includes the sending clock frequency of the SoC domain in the first die.

[0174] 707 . The first die stops sending the associated clock to the second die, and adjusts the operating frequency of the sending circuit of the first die according to the frequency level.

[0175] The clock of the transmitting circuit in the first die can be connected to the clock of the receiving circuit in the second die via a clock line. The clock of the receiving circuit in the second die is a slave clock to the clock of the transmitting circuit in the first die. When the clock frequency of the transmitting circuit in the first die changes, the slave clock of the receiving circuit in the second die also changes accordingly.

[0176] Adjusting the operating frequency of the transmitting circuit of the first bare chip according to the frequency level by the first bare chip can be understood as follows: the first bare chip determines the circuit configuration parameters at the new frequency level according to the training parameters pre-stored in the interconnection interface circuit of the first bare chip, that is, the correspondence between the frequency level and the circuit configuration parameters, so as to configure the circuit parameters of the first bare chip according to the circuit configuration parameters, so that the circuit parameters of the configured first bare chip adapt to the new frequency level.

[0177] The circuit configuration parameters in the training parameters of the first bare die may be understood as circuit configuration parameters of a transmitting circuit in the interconnection interface circuit of the first bare die.

[0178] In some embodiments, the training parameters may be pre-stored when the die is shipped from the factory.

[0179] 708. The first die sends a frequency switching instruction to the second die.

[0180] In some embodiments, the first die may send a frequency switching instruction to the second die via a sideband signal command (SBC_SYNC_IND), where the frequency switching instruction is used to instruct the second die to perform frequency switching.

[0181] The format of the sideband signal command can be seen in Table 3 and Table 5. In the sideband signal command, the value of the type[2:0] field can be "111", and the value of the opcode[7:0] field can be "11111111".

[0182] 709. When the second bare chip receives the frequency switching instruction, the second bare chip adjusts the operating frequency of the receiving circuit of the second bare chip according to the frequency level.

[0183] The second bare chip determines circuit configuration parameters according to the information indicating the frequency level after the switch, and configures circuit parameters of a receiving circuit in the interconnection interface circuit of the second bare chip according to the circuit configuration parameters.

[0184] Similarly, the second bare chip adjusts the operating frequency of the transmitting circuit of the second bare chip according to the frequency level, which can be understood as the second bare chip determining the circuit configuration parameters at the new frequency level based on the training parameters pre-stored in the interconnection interface circuit of the second bare chip, that is, the correspondence between the frequency level and the circuit configuration parameters, so as to configure the circuit parameters of the receiving circuit of the second bare chip according to the circuit configuration parameters, so that the circuit parameters of the receiving circuit of the second bare chip after configuration adapt to the new frequency level.

[0185] In some embodiments, the training parameters of the first die and the training parameters of the second die may be the same or different. When the training parameters of the first die and the training parameters of the second die are different, it can be understood that the training parameters of the first die and the training parameters of the second die have the same frequency levels, but different circuit configuration parameters at the same frequency level.

[0186] In some embodiments, when the training parameters of the first die and the training parameters of the second die are the same, the first die may send the stored training parameters to the second die when the first die and the second die are powered on and initialized.

[0187] In some embodiments, the first die may complete circuit parameter configuration before the first die sends a frequency switch request to the second die, or while the first die is sending the frequency switch request to the second die, or after the first die receives a frequency switch response, or after the application layer switches the clock frequency of the SoC domain. That is, the time when the first die and the second die complete circuit parameter configuration may not be synchronized.

[0188] 710. After the first die waits for a fixed delay, the associated clock of the first die is started.

[0189] The fixed delay needs to cover the stabilization time after the second die switches the frequency and the stabilization time after the first die switches the frequency.

[0190] 711. The first die sends data to the second die at the frequency level.

[0191] Accordingly, the second die receives the data sent by the first die at the frequency level.

[0192] In some embodiments, when the first die receives an indication from the second die that circuit parameter configuration is complete, and when the first die also completes circuit parameter configuration, the first die may send service data to the second die at the switched frequency level.

[0193] Therefore, in the frequency modulation method of the present application, when the SoC domain and the interconnection interface are in the same clock domain, the training parameters of the interconnection interface of the bare chip can be pre-stored in the circuit of the bare chip, and the indication information of the frequency level after interactive switching through the high-speed data channel or the sideband control channel can be used. Both bare chips can call the stored training parameters according to the frequency level to complete the rapid switching of the frequency level.

[0194] In addition, in the present application, considering that it is necessary to stop receiving service data from the application layer during the frequency switching process and keep the service flow uninterrupted, when fast switching at the frequency level is achieved, it is beneficial to reduce the cache of service data.

[0195] When the sending and receiving of service data is stopped during the frequency switching process, the switching processing delay can be reduced when implementing fast switching at the frequency level, which is equivalent to reducing the back pressure time for service data.

[0196] Furthermore, when the SoC domain of the first die, the transmit circuitry of the first die, and the receive circuitry of the second die are in the same clock domain, the digital logic of the digital link layer of the first die can dynamically adjust its voltage based on the frequency changes of the SoC domain, implementing dynamic voltage and frequency scaling (DVFS), which can further reduce the power consumption of the first die. Similarly, the power consumption of the second die can also be reduced.

[0197] It is understandable that, in order to realize the above functions, the above-mentioned bare chip includes hardware structures and / or software modules corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments of the present application.

[0198] In the embodiment of the present application, the above-mentioned bare chip can be divided into functional modules according to the above-mentioned method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. In actual implementation, other division methods can be used.

[0199] In the case of dividing the functional modules according to their respective functions, FIG11 shows a possible composition diagram of the bare chip 110 involved in the above embodiment. As shown in FIG11 , the bare chip 110 may include: a sending unit 1101 , an adjustment unit 1102 and a receiving unit 1103 .

[0200] The transmitting unit 1101 can be used to implement the function of the transmitting circuit of the interconnect interface in the above-mentioned bare chip, and the receiving unit 1103 can be used to implement the transmitting function of the receiving circuit in the interconnect interface in the above-mentioned bare chip. The adjusting unit 1102 can be used to implement the function of switching the operating frequency of the SoC domain, the operating frequency of the transmitting circuit, and the operating frequency of the receiving circuit in the above-mentioned bare chip.

[0201] The sending unit 1101 may be used to support the die 110 in executing the above steps 501 , 702 , 704 , 705 , 708 , 711 , etc., and / or other processes of the technology described herein.

[0202] The adjustment unit 1102 may be configured to support the die 110 in executing the aforementioned steps 502 , 703 , 706 , 707 , 709 , 710 , etc., and / or other processes for the technology described herein.

[0203] The receiving unit 1103 may be configured to support the die 110 in executing the aforementioned step 701 and / or other processes of the technology described herein.

[0204] It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0205] The bare chip 110 provided in this embodiment is used to execute the above-mentioned frequency adjustment method, and thus can achieve the same effect as the above-mentioned implementation method.

[0206] In the case of adopting an integrated unit, as shown in Figure 12, an embodiment of the present application discloses a bare chip 120, which can be the chip system in the above embodiment, such as a small chip / core particle of a chiplet system. The bare chip 120 may include a processing module, a storage module and a communication module. Among them, the processing module can be used to control and manage the actions of the bare chip 120, for example, it can be used to support the bare chip 120 to execute the process of the above-mentioned adjustment unit 1102. The storage module can be used to support the bare chip 120 to store program code and data, etc. For example, the storage module stores the training data in this application, including circuit configuration parameters corresponding to the frequency level. The communication module can be used to support communication between the bare chip 120 and other bare chips. The communication module may include the interconnection interface in this application.

[0207] Of course, the unit modules in the bare chip 120 include but are not limited to the processing module, storage module and communication module. For example, the bare chip 120 may also include a power module, etc. The power module is used to supply power to the bare chip 120.

[0208] The processing module may be a processor or a controller. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processing (DSP) and a microprocessor, and so on. The storage module may be a memory. The communication module may specifically be a device such as a chip interface that interacts with other external devices.

[0209] For example, the processing module may be processor 1301, the storage module may be memory 1302, and the communication module may be referred to as communication interface 1303, such as the interconnect interface in this application. The bare die 120 provided in this embodiment of the application may be chip 130 shown in FIG13. The processor, memory, communication interface, etc. may be connected together, for example, via a bus.

[0210] An embodiment of the present application further provides a communication device comprising one or more processors and one or more memories. The one or more memories are coupled to the one or more processors and are configured to store computer program code. The computer program code comprises computer instructions. When the one or more processors execute the computer instructions, the communication device executes the aforementioned related method steps to implement the frequency modulation method of the aforementioned embodiment.

[0211] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program code is stored. When a processor executes the computer program code, the communication device executes the frequency modulation method in the above embodiment.

[0212] An embodiment of the present application further provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the frequency modulation method executed by the communication device in the above-mentioned embodiment.

[0213] Among them, the communication device provided in this embodiment, such as a bare chip, a computer storage medium, a computer program product or a chip, is used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be repeated here.

[0214] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0215] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0216] The units described as separate components may or may not be physically separate, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0217] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0218] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0219] The above content is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A frequency modulation method, characterized in that: The method is applied to a chip system, the chip system including a first die and a second die, and the method includes: The first die sends a frequency switching request to the second die, the frequency switching request including information indicating a frequency level after switching; The second die adjusts an operating frequency of a receiving circuit of the second die according to the frequency level.

2. The method according to claim 1, characterized in that The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

3. The method according to claim 1 or 2, characterized in that Before the first die sends the frequency switching request to the second die, the method further includes: The first bare chip receives a frequency mode configuration command sent by the application layer, where the frequency mode configuration command includes first indication information, and the first indication information is used to indicate the frequency level after the switch; The first die generates the frequency switching request according to the frequency mode configuration command.

4. The method according to any one of claims 1 to 3, characterized in that Before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: The first die stops receiving service data from the application layer.

5. The method according to any one of claims 1 to 4, characterized in that The first die sending a frequency switching request to the second die includes: The first die sends the frequency switching request to the second die through a first low power control word LPW flow control unit or a first sideband control command; The method further comprises: The first die receives the frequency switching response sent by the second die through a second LPW flow control unit or a second sideband control command.

6. The method according to claim 5, characterized in that The first LPW flow control unit includes a flow control unit header field, a flow control unit content field, and a flow control unit tail field; The flow control unit content field includes a type subfield and a command subfield. The type subfield is used to indicate that the first LPW flow control unit is a frequency switching command, and the command subfield is used to indicate the frequency level.

7. The method according to claim 5, characterized in that The first sideband control command includes a field indicating a message type, a field indicating a message code, a field indicating a path identifier, and a field indicating a check bit; The field indicating the message type is used to indicate that the first sideband control command is a frequency switching command.

8. The method according to any one of claims 1 to 7, characterized in that Before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: The first die stops sending the associated clock to the second die, and adjusts the operating frequency of the transmitting circuit of the first die according to the frequency level.

9. The method according to any one of claims 1 to 8, characterized in that Before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: The first die continues to send empty flow control units to the second die until receiving a frequency switching response sent by the second die.

10. The method according to any one of claims 1 to 9, characterized in that Before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: After the first bare chip determines that the frequency switching negotiation with the second bare chip is completed, the first bare chip responds to the application layer that the frequency switching negotiation is completed.

11. The method according to any one of claims 1 to 10, characterized in that Before the second die adjusts the operating frequency of the receiving circuit of the second die, the method further includes: The first die sends a frequency switching instruction to the second die, where the frequency switching instruction is used to instruct the second die to perform frequency switching.

12. The method according to any one of claims 1 to 11, characterized in that The method further comprises: After the first die waits for a fixed delay, the associated clock of the first die is started.

13. The method according to any one of claims 1 to 12, characterized in that The method further comprises: The first die sends data to the second die at the frequency level; The second die receives the data sent by the first die at the frequency level.

14. A frequency modulation method, characterized in that: The method is applied to a chip system, the chip system including a first die and a second die, and the method includes: The second bare chip receives the frequency switching request sent by the first bare chip, where the frequency switching request includes information indicating a frequency level after switching; The second die adjusts an operating frequency of a receiving circuit of the second die according to the frequency level.

15. The method according to claim 14, characterized in that The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

16. The method according to claim 14 or 15, characterized in that The second bare chip receiving the frequency switching request sent by the first bare chip includes: The second die receives the frequency switching request sent by the first die through a first low power control word LPW flow control unit or a first sideband control command; The method further comprises: The second die sends a frequency switching response to the first die through a second LPW flow control unit or a second sideband control command.

17. The method according to any one of claims 15-16, characterized in that The method further comprises: When the second bare chip receives the frequency switching request sent by the first bare chip, the second bare chip records the information indicating the frequency level after switching.

18. The method according to any one of claims 15 to 17, characterized in that: Before the second die adjusts the operating frequency of the receiving circuit of the second die according to the frequency level, the method further includes: The second bare chip receives a frequency switching instruction sent by the first bare chip, where the frequency switching instruction is used to instruct the second bare chip to perform frequency switching.

19. A frequency modulation method, characterized in that: The method is applied to a chip system, the chip system including a first die and a second die, and the method includes: The first die sends a frequency switching request to the second die, the frequency switching request including information indicating a frequency level after switching; The first die adjusts an operating frequency of a transmitting circuit of the first die according to the frequency level.

20. The method according to claim 19, characterized in that The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

21. The method according to claim 19 or 20, characterized in that The first die sending a frequency switching request to the second die includes: The first die sends the frequency switching request to the second die through a first low power control word LPW flow control unit or a first sideband control command; The method further comprises: The first die receives the frequency switching response sent by the second die through a second LPW flow control unit or a second sideband control command.

22. A chip system, characterized in that: The chip system includes a first die and a second die, wherein: The first bare chip is configured to send a frequency switching request to the second bare chip, wherein the frequency switching request includes information indicating a frequency level after switching; The second die is configured to adjust an operating frequency of a receiving circuit of the second die according to the frequency level.

23. The chip system according to claim 22, characterized in that: The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

24. The chip system according to claim 22 or 23, characterized in that: The first die is used for: receiving a frequency mode configuration command sent by an application layer, where the frequency mode configuration command includes first indication information, where the first indication information is used to indicate the frequency level after the switch; The frequency switching request is generated according to the frequency mode configuration command.

25. The chip system according to any one of claims 22 to 24, characterized in that: The first die is used for: Sending the frequency switching request to the second bare chip through a first low power control word LPW flow control unit or a first sideband control command; A frequency switching response sent by the second die is received through a second LPW flow control unit or a second sideband control command.

26. A second bare chip, characterized in that: The chip system includes a first die and a second die, wherein the second die includes: a receiver, configured to receive a frequency switching request sent by the first bare chip, wherein the frequency switching request includes information indicating a frequency level after switching; The processor is configured to adjust an operating frequency of a receiving circuit of the second die according to the frequency level.

27. The second die according to claim 26, wherein: The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

28. The second die according to claim 26 or 27, wherein: The receiver is configured to: Receiving the frequency switching request sent by the first bare chip through a first low power control word LPW flow control unit or a first sideband control command; The receiver is further configured to: A frequency switching response is sent to the first die through a second LPW flow control unit or a second sideband control command.

29. The second die according to any one of claims 26 to 28, wherein: The processor is further configured to: The receiver is configured to record the information indicating the frequency level after switching when receiving the frequency switching request sent by the first bare chip.

30. The second die according to any one of claims 26 to 29, wherein: The receiver is further configured to receive a frequency switching instruction sent by the first bare chip, where the frequency switching instruction is used to instruct the second bare chip to perform frequency switching.

31. A first bare chip, characterized in that: The chip system includes the first die and the second die, wherein the first die includes: a transmitter, configured to send a frequency switching request to the second bare chip, wherein the frequency switching request includes information indicating a frequency level after switching; The processor is configured to adjust an operating frequency of a transmitting circuit of the first die according to the frequency level.

32. The first die according to claim 31, wherein: The transmitting circuit of the first die and the receiving circuit of the second die are in the same clock domain.

33. The first bare chip according to claim 31 or 32, characterized in that The transmitter is used to: Sending the frequency switching request to the second bare chip through a first low power control word LPW flow control unit or a first sideband control command; The system further includes a receiver configured to receive a frequency switching response sent by the second die through a second LPW flow control unit or a second sideband control command.

34. A computer-readable storage medium, characterized in that The method comprises computer instructions, which, when executed on an electronic device, cause the electronic device to execute the method according to any one of claims 1 to 13.

35. A computer-readable storage medium, characterized in that The method comprises computer instructions, which, when executed on an electronic device, cause the electronic device to execute the method according to any one of claims 14 to 18.

36. A computer-readable storage medium, characterized in that The method comprises computer instructions, which, when executed on an electronic device, cause the electronic device to execute the method according to any one of claims 19 to 21.