Chip global layout method and device, electronic device, and readable storage medium

Optimizing the chip layout through the neural network model and DREAMPlace method solves the problems of slow speed and high computing cost in large-scale integrated circuits, and achieves more efficient layout optimization and performance improvement.

CN117408216BActive Publication Date: 2024-11-12BEIJING UNIV OF POSTS & TELECOMM
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202311452203.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-03
Publication Date
2024-11-12
Estimated Expiration
2043-11-03

AI Technical Summary

Technical Problem

Traditional chip layout algorithms cannot effectively handle the layout of a large number of circuit modules in large-scale integrated circuits, resulting in slow speed and high calculation costs, and the optimal solution cannot be obtained.

Method used

Using a neural network model trained based on preset circuit features, the circuit characteristics of macrocells and standard units are extracted, their positions are optimized to obtain global layout positions, and the layout is optimized through the Decision Transformer and DREAMPlace methods to ensure no overlap and the overall wiring length is optimal.

Benefits of technology

It improves the efficiency and effect of chip layout, reduces computing costs, ensures the optimization of wiring length between macro cells and standard cells, and improves overall layout performance and power consumption management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117408216B_ABST
    Figure CN117408216B_ABST
Patent Text Reader

Abstract

The present disclosure provides a chip global layout method and device, electronic device, and readable storage medium, the method comprising: extracting the circuit characteristics of the macro unit in the first chip according to the circuit netlist of the first chip to obtain the first circuit characteristics; inputting the first circuit characteristics into a preset chip layout model to obtain the first layout position of the macro unit; wherein the preset chip layout model is a model trained based on the preset circuit characteristics; optimizing the positions of the macro unit and the standard unit in the first chip based on the first layout position to obtain the global layout position of the first chip. The chip global layout method and device, electronic device, and readable storage medium provided by the present disclosure can improve the effect of chip layout.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure belongs to the field of data processing technology, and more specifically, relates to a chip global layout method and device, an electronic device, and a readable storage medium. Background Art

[0002] Chip layout is the process of placing circuit modules of different sizes (including macro cells and standard cells) on the chip canvas. The chip layout result determines the performance and cost of the chip and is a crucial and challenging issue in chip design.

[0003] As the scale of current integrated circuits continues to grow, traditional chip layout algorithms are unable to cope with the number of layout units in current large-scale integrated circuits. They are slow and have high computational costs, and cannot obtain the optimal solution. Summary of the invention

[0004] The purpose of the present disclosure is to provide a chip global layout method and device, an electronic device, and a readable storage medium to improve the effect of chip layout.

[0005] According to a first aspect of an embodiment of the present disclosure, a chip global layout method is provided, comprising:

[0006] Extracting circuit characteristics of the macro unit in the first chip according to the circuit netlist of the first chip to obtain first circuit characteristics;

[0007] Inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; wherein the preset chip layout model is a model trained based on preset circuit features;

[0008] The positions of macro cells and standard cells in the first chip are optimized based on the first layout position to obtain a global layout position of the first chip.

[0009] A second aspect of the embodiments of the present disclosure provides a chip global layout device, including:

[0010] A feature extraction unit, configured to extract circuit features of the macrocell in the first chip according to a circuit netlist of the first chip to obtain first circuit features;

[0011] A position calculation unit, used for inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; wherein the preset chip layout model is a model trained based on preset circuit features;

[0012] A position optimization unit optimizes the positions of macro cells and standard cells in the first chip based on the first layout position to obtain a global layout position of the first chip.

[0013] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned chip global layout method when executing the computer program.

[0014] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned chip global layout method are implemented.

[0015] The chip global layout method and device, electronic device, and readable storage medium provided by the embodiments of the present disclosure have the following beneficial effects:

[0016] The disclosed embodiment first obtains a first layout position of a macro cell in a first chip by training a preset chip layout model, and then optimizes the positions of the macro cell and the standard cell in the first chip based on the first layout position, and finally obtains a global layout position.

[0017] Since the macrocell occupies a large area in the first chip, it has a great impact on the total wiring length on the first chip, and the total wiring length of the circuit affects the performance, power consumption, manufacturing cost, etc. of the chip. First, the layout position of the macrocell is optimized to ensure that there is no overlap between the macrocells, and the wiring length between the macrocells is optimized, and then the layout position of the standard cell is optimized to ensure that there is no overlap as a whole and the total wiring length is controllable, so as to achieve the purpose of improving the overall layout effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of a process flow of a chip global layout method provided by an embodiment of the present disclosure;

[0020] Figure 2 A schematic diagram of a chip on-chip unit diagram provided by an embodiment of the present disclosure;

[0021] Figure 3 A structural block diagram of a chip global layout device provided by an embodiment of the present disclosure;

[0022] Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0024] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0025] Please refer to Figure 1 , Figure 1 A schematic diagram of a chip global layout method provided in an embodiment of the present disclosure, the method comprising:

[0026] S101: Extracting circuit characteristics of a macro cell in a first chip according to a circuit netlist of a first chip to obtain first circuit characteristics.

[0027] In the embodiment of the present disclosure, the first chip is a chip to be laid out, and the circuit netlist includes various on-chip units (including macro units and standard units) in the chip circuit and the connection relationship between the various on-chip units.

[0028] The circuit netlist is obtained based on the on-chip unit diagram. A schematic diagram of the on-chip unit diagram is shown as follows: Figure 2 shown. Figure 2 In the figure, each rectangular block represents a macro cell or a standard cell, and the connection lines between the rectangular blocks represent the connection relationship between the macro cells and / or the standard cells.

[0029] Those skilled in the art can extract the circuit features of the macro cells in the first chip from the circuit netlist using existing means, which will not be elaborated here.

[0030] S102: Inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; wherein the preset chip layout model is a model trained based on the preset circuit feature.

[0031] In the disclosed embodiment, the global layout data of an existing chip may be used. The global layout data includes a circuit netlist of the existing chip and the position coordinates of all layout macro units. Preset circuit features are extracted based on the circuit netlist of the existing chip.

[0032] Furthermore, the preset circuit features can be used as input data, and the position coordinates of the corresponding macro units can be used as output data to train the neural network model to obtain the preset chip layout model. In the disclosed embodiment, the neural network model specifically adopts an offline reinforcement learning model based on Decision Transformer.

[0033] On this basis, the first circuit feature is input into the trained neural network model to obtain the first layout position of the macro unit.

[0034] S103: Optimizing the positions of macro cells and standard cells in the first chip based on the first layout position to obtain a global layout position of the first chip.

[0035] In the embodiment of the present disclosure, on the basis of obtaining the first layout position of the macro cell, the positions of the macro cell and the standard cell are further optimized to obtain the global layout position of the first chip.

[0036] From the above, it can be concluded that the embodiment of the present disclosure first obtains the first layout position of the macro unit in the chip to be laid out by training the neural network model, and then uses the first layout position of the macro unit as the initial layout position to synchronously optimize the macro unit and the standard unit in the chip to be laid out, and finally obtains the global layout optimization result.

[0037] Since macro cells occupy a large area in the chip to be laid out, they have a great impact on the total wiring length on the chip to be laid out, and the total wiring length of the circuit affects the performance, power consumption, manufacturing cost, etc. of the chip. First, optimize the layout position of the macro cells to ensure that there is no overlap between macro cells, and optimize the wiring length between macro cells, and then optimize the layout position of the standard cells to ensure that there is no overlap as a whole and the total wiring length is controllable, so as to achieve the purpose of improving the overall layout effect.

[0038] In one embodiment of the present disclosure, inputting a first circuit feature into a preset chip layout model to obtain a first layout position of a macro unit includes:

[0039] In response to the similarity between the first circuit feature and the preset circuit feature being greater than or equal to the preset similarity, the first circuit feature is input into a preset chip layout model to obtain a first layout position of the macro unit.

[0040] In the disclosed embodiment, the preset chip layout model is obtained by training according to the preset circuit feature. The greater the similarity between the first circuit feature and the preset circuit feature, the more likely the layout position of the macro unit obtained according to the preset chip layout model is the optimal layout position. Therefore, when the similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, the layout position obtained by the preset chip layout model can be directly used as the first layout position without subsequent verification, which is conducive to reducing the amount of calculation and improving the calculation speed.

[0041] In one embodiment of the present disclosure, the first circuit feature is input into a preset chip layout model to obtain a first layout position of a macro unit, and further includes:

[0042] In response to the similarity between the first circuit feature and the preset circuit feature being less than the preset similarity, the parameters of the preset chip layout model are adjusted, and the first layout position is obtained according to the adjusted preset chip layout model.

[0043] In an embodiment of the present disclosure, when the similarity between the first circuit feature and the preset circuit feature is less than the preset similarity, the preset chip layout model is fine-tuned by adjusting the preset chip layout model parameters, and the first layout position is obtained according to the fine-tuned preset chip layout model to adapt to the new chip on-chip environment.

[0044] In one embodiment of the present disclosure, adjusting parameters of a preset chip layout model, and obtaining a first layout position according to the adjusted preset chip layout model, includes:

[0045] Adjust the parameters of the preset chip layout model multiple times.

[0046] After each parameter adjustment, the first circuit feature is input into the adjusted preset chip layout model to obtain a second layout position.

[0047] The optimal position among the plurality of second layout positions is selected, and the optimal position is determined as the first layout position. The optimal position is the position where the connection length between the macro units is the shortest.

[0048] In the disclosed embodiment, the parameters of the preset chip layout model can be adjusted multiple times to perform online reinforcement learning exploration on the first chip. Through multiple layout explorations, the optimal layout result (i.e., no unit overlap and the total line length between units is the shortest) is retained as the final global layout result.

[0049] Furthermore, the global layout data of the first chip and the first layout position may be added to a training set of a preset chip layout model, and the preset chip layout model may be retrained to improve the versatility of the preset chip layout model.

[0050] In one embodiment of the present disclosure, the first circuit feature includes a first adjacency matrix and a first node matrix. The preset circuit feature includes a preset adjacency matrix and a preset node matrix.

[0051] The similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, including:

[0052] The similarity between the first adjacency matrix and the preset adjacency matrix is ​​greater than or equal to a first similarity, and the similarity between the first node matrix and the preset node matrix is ​​greater than or equal to a second similarity.

[0053] In the disclosed embodiment, each on-chip unit (including macro units and standard units) in the circuit netlist is used as a node, and the connection relationship between each on-chip unit is used as the relationship between nodes. The corresponding adjacency matrix and node matrix are extracted using a graph neural network (GNN), and the adjacency matrix and node matrix are used as the circuit features of the chip. According to the above method, the first adjacency matrix and the first node matrix, as well as the preset adjacency matrix and the preset node matrix are obtained respectively.

[0054] When judging whether the similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, the first adjacency matrix is ​​compared with the preset adjacency matrix, and the first node matrix is ​​compared with the preset node matrix. Only when the similarity between the first adjacency matrix and the preset adjacency matrix is ​​greater than or equal to the first similarity, and the similarity between the first node matrix and the preset node matrix is ​​greater than or equal to the second similarity, is it judged that the similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity. In the disclosed embodiment, the similarity between the two matrices is calculated by calculating the Euclidean distance between the two matrices.

[0055] Those skilled in the art can flexibly design specific values ​​of the preset similarity, the first similarity and the second similarity, such as 95%, 99%, etc., which are not limited here.

[0056] In one embodiment of the present disclosure, the first circuit feature includes a first adjacency matrix and a first node matrix. The preset circuit feature includes a preset adjacency matrix and a preset node matrix.

[0057] The similarity between the first circuit feature and the preset circuit feature is less than the preset similarity, including:

[0058] The similarity between the first adjacency matrix and the preset adjacency matrix is ​​smaller than the first similarity, or the similarity between the first node matrix and the preset node matrix is ​​smaller than the second similarity.

[0059] In the embodiment of the present disclosure, if the similarity between the first adjacency matrix and the preset adjacency matrix is ​​less than the first similarity, or the similarity between the first node matrix and the preset node matrix is ​​less than the second similarity, it is determined that the similarity between the first circuit feature and the preset circuit feature is less than the preset similarity.

[0060] In one embodiment of the present disclosure, optimizing the positions of macro cells and standard cells in a first chip based on a first layout position to obtain a global layout position of the first chip includes:

[0061] The first layout position is determined as the initial position in the DREAMPlace method, and the positions of the macro cells and the standard cells in the first chip are optimized based on the DREAMPlace method to obtain the global layout position of the first chip.

[0062] In the disclosed embodiment, the DREAMPlace method uses the gradient descent algorithm of pytorch to model each on-chip unit as an electrostatic charge, uses electrostatic repulsion instead of unit overlap, and optimizes overlap and line length at the same time to optimize the positions of macro units and standard units in the first chip. Since the DREAMPlace method uses a gradient descent optimization algorithm, the initial position of each on-chip unit has an impact on the final optimization result. The first layout position is used as the initial position of the macro unit to start the optimization. Compared with the optimization from a completely random initial position, it not only improves the optimization speed, but also improves the optimization effect.

[0063] Corresponding to the chip global layout method of the above embodiment, Figure 3 This is a block diagram of a chip global layout device provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 3 The chip global layout device 20 includes: a feature extraction unit 21, a position calculation unit 22, and a position optimization unit 23.

[0064] The feature extraction unit 21 is used to extract the circuit features of the macro cells in the first chip according to the circuit netlist of the first chip to obtain the first circuit features.

[0065] The position calculation unit 22 is used to input the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit, wherein the preset chip layout model is a model trained based on the preset circuit feature.

[0066] The position optimization unit 23 optimizes the positions of the macro cells and the standard cells in the first chip based on the first layout position to obtain the global layout position of the first chip.

[0067] In one embodiment of the present disclosure, the position calculation unit 22 is specifically used for:

[0068] In response to the similarity between the first circuit feature and the preset circuit feature being greater than or equal to the preset similarity, the first circuit feature is input into a preset chip layout model to obtain a first layout position of the macro unit.

[0069] In one embodiment of the present disclosure, the position calculation unit 22 is further used for:

[0070] In response to the similarity between the first circuit feature and the preset circuit feature being less than the preset similarity, the parameters of the preset chip layout model are adjusted, and the first layout position is obtained according to the adjusted preset chip layout model.

[0071] In one embodiment of the present disclosure, the position calculation unit 22 is further used for:

[0072] Adjust the parameters of the preset chip layout model multiple times.

[0073] After each parameter adjustment, the first circuit feature is input into the adjusted preset chip layout model to obtain a second layout position.

[0074] The optimal position among the plurality of second layout positions is selected, and the optimal position is determined as the first layout position. The optimal position is the position where the connection length between the macro units is the shortest.

[0075] In one embodiment of the present disclosure, the similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, including:

[0076] The similarity between the first adjacency matrix and the preset adjacency matrix is ​​greater than or equal to a first similarity, and the similarity between the first node matrix and the preset node matrix is ​​greater than or equal to a second similarity.

[0077] In one embodiment of the present disclosure, the similarity between the first circuit feature and the preset circuit feature is less than the preset similarity, including:

[0078] The similarity between the first adjacency matrix and the preset adjacency matrix is ​​smaller than the first similarity, or the similarity between the first node matrix and the preset node matrix is ​​smaller than the second similarity.

[0079] In one embodiment of the present disclosure, the location optimization unit 23 is specifically used to:

[0080] The first layout position is determined as the initial position in the DREAMPlace method, and the positions of the macro cells and the standard cells in the first chip are optimized based on the DREAMPlace method to obtain the global layout position of the first chip.

[0081] See also Figure 4 , Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 4The terminal 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to perform the following operations to perform the functions of the modules / units in the above-mentioned device embodiments, such as Figure 3 The functions of modules 21 to 23 are shown.

[0082] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0083] The input device 302 may include a touch panel, a fingerprint sensor (for collecting fingerprint information of the user and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0084] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0085] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure may execute the implementation methods described in the first and second embodiments of the chip global layout method provided in the embodiments of the present disclosure, and may also execute the implementation methods of the terminal described in the embodiments of the present disclosure, which will not be repeated here.

[0086] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, wherein the program instructions are executed by a processor to implement all or part of the processes in the above-mentioned embodiment method, and may also be completed by instructing the relevant hardware through the computer program, wherein the computer program may be stored in a computer-readable storage medium, and wherein the computer program may be executed by the processor to implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the contents contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signal and telecommunication signal.

[0087] The computer-readable storage medium may be an internal storage unit of the terminal of any of the foregoing embodiments, such as a hard disk or memory of the terminal. The computer-readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0088] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the terminals and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0090] In the several embodiments provided in the present application, it should be understood that the disclosed terminals and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0091] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0092] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0093] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A global chip layout method, It is characterized in that include: Extracting circuit characteristics of the macro unit in the first chip according to the circuit netlist of the first chip to obtain first circuit characteristics; Inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; wherein the preset chip layout model is a model trained based on preset circuit features; Optimizing the positions of macro cells and standard cells in the first chip based on the first layout position to obtain a global layout position of the first chip; The step of inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit includes: In response to the similarity between the first circuit feature and a preset circuit feature being greater than or equal to a preset similarity, inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; The first circuit feature includes a first adjacency matrix and a first node matrix; the preset circuit feature includes a preset adjacency matrix and a preset node matrix; wherein, each on-chip unit in the circuit netlist of the first chip is taken as a node, and the connection relationship between each on-chip unit is taken as the relationship between nodes, and the corresponding adjacency matrix and node matrix are extracted using a graph neural network to obtain a first adjacency matrix and a first node matrix; each on-chip unit in the circuit netlist of an existing chip is taken as a node, and the connection relationship between each on-chip unit is taken as the relationship between nodes, and the corresponding adjacency matrix and node matrix are extracted using a graph neural network to obtain a preset adjacency matrix and a preset node matrix; The similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, including: The similarity between the first adjacency matrix and the preset adjacency matrix is ​​greater than or equal to a first similarity, and the similarity between the first node matrix and the preset node matrix is ​​greater than or equal to a second similarity.

2. The chip global layout method according to claim 1, It is characterized in that The step of inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit further includes: In response to the similarity between the first circuit feature and the preset circuit feature being less than a preset similarity, a parameter of the preset chip layout model is adjusted, and a first layout position is obtained according to the adjusted preset chip layout model.

3. The chip global layout method according to claim 2, It is characterized in that The step of adjusting the parameters of the preset chip layout model and obtaining the first layout position according to the adjusted preset chip layout model includes: Adjusting the parameters of the preset chip layout model multiple times; After each parameter adjustment, the first circuit feature is input into the adjusted preset chip layout model to obtain a second layout position; An optimal position among the plurality of second layout positions is selected, and the optimal position is determined as the first layout position; the optimal position is a position where the connection length between the macro units is the shortest.

4. The chip global layout method according to claim 2, It is characterized in that The similarity between the first circuit feature and the preset circuit feature is less than the preset similarity, comprising: The similarity between the first adjacency matrix and the preset adjacency matrix is ​​smaller than the first similarity, or the similarity between the first node matrix and the preset node matrix is ​​smaller than the second similarity.

5. The chip global layout method according to claim 1, It is characterized in that The optimizing the positions of the macro cells and the standard cells in the first chip based on the first layout position to obtain the global layout position of the first chip includes: The first layout position is determined as an initial position in a DREAMPlace method, and the positions of macro cells and standard cells in the first chip are optimized based on the DREAMPlace method to obtain a global layout position of the first chip.

6. A chip global layout device, It is characterized in that include: A feature extraction unit, configured to extract circuit features of the macrocell in the first chip according to a circuit netlist of the first chip to obtain first circuit features; A position calculation unit, used for inputting the first circuit feature into a preset chip layout model to obtain a first layout position of the macro unit; wherein the preset chip layout model is a model trained based on preset circuit features; A position optimization unit, which optimizes the positions of macro cells and standard cells in the first chip based on the first layout position to obtain a global layout position of the first chip; The position calculation unit is specifically used for: In response to the similarity between the first circuit feature and the preset circuit feature being greater than or equal to the preset similarity, the first circuit feature is input into a preset chip layout model to obtain a first layout position of the macro unit; The first circuit feature includes a first adjacency matrix and a first node matrix; the preset circuit feature includes a preset adjacency matrix and a preset node matrix; wherein, each on-chip unit in the circuit netlist of the first chip is taken as a node, and the connection relationship between each on-chip unit is taken as the relationship between nodes, and the corresponding adjacency matrix and node matrix are extracted using a graph neural network to obtain a first adjacency matrix and a first node matrix; each on-chip unit in the circuit netlist of an existing chip is taken as a node, and the connection relationship between each on-chip unit is taken as the relationship between nodes, and the corresponding adjacency matrix and node matrix are extracted using a graph neural network to obtain a preset adjacency matrix and a preset node matrix; The similarity between the first circuit feature and the preset circuit feature is greater than or equal to the preset similarity, including: The similarity between the first adjacency matrix and the preset adjacency matrix is ​​greater than or equal to a first similarity, and the similarity between the first node matrix and the preset node matrix is ​​greater than or equal to a second similarity.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Citation Information

Patent Citations

  • Test method for physical design similarity of circuit

    CN103020346A

  • Search method, search device, and search system

    CN113826101A

  • Circuit board part installation detection method

    CN114187253A

  • Graph neural network training method and device, equipment and storage medium

    CN114330670A

  • Integrated circuit layout optimization method based on reinforcement learning

    CN114896937A