Clustering-based H-shaped clock tree trunk node coordinate selection method and system
By automatically selecting the backbone node coordinates of the H-shaped clock tree based on clustering, the problem of inefficient construction of H-shaped clock tree in digital integrated circuit design is solved, and the rapid and efficient construction of H-shaped clock tree is realized, which is automatic and flexible Advantages of good sexual and adaptability.
Patent Information
- Application Number
- CN202210487931.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-06
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-05-06
AI Technical Summary
In digital integrated circuit design, the prior art lacks automated methods to select the backbone node coordinates of the H-shaped clock tree, resulting in inefficient construction of the H-shaped clock tree.
Using a cluster-based method, by obtaining the triggers in a given area, initializing the coordinates of the main nodes, and assigning the triggers to the corresponding cluster, calculating the coordinates of the cluster center point, and automatically selecting the main nodes of the H-shaped clock tree. Avoid manual specification.
An H-shaped clock tree structure with high automation, flexibility and adaptability is realized, and the H-shaped clock tree in a given area is quickly and efficiently generated, avoiding the arbitraryness and inefficiency of manual designation.
Smart Images

Figure CN114880983B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to a very large scale integrated circuit clock tree design technology, and in particular relates to a clustering-based H-shaped clock tree trunk node coordinate selection method and system. Background Art
[0002] In digital integrated circuit design, the clock signal is the benchmark for data transmission. It plays a decisive role in the function, performance and stability of synchronous digital systems. Therefore, the characteristics of the clock signal and the layout of the clock tree are particularly important. At present, the mainstream clock tree structures of digital integrated circuits are H-tree, X-tree, balanced-tree and clock mesh. H-Tree is named because its physical shape is similar to the letter H. Its main feature is that each level of driver is physically connected through an H-shaped metal network, which gradually expands from the clock root node and covers all leaf nodes. The distance from the root node to each leaf node of H-Tree is theoretically equal, so it has the advantage of small clock deviation and is widely used in the current physical design field. In order to quickly construct an H-shaped clock tree, electronic design automation (EDA) tools usually require physical designers to specify the coordinates of the trunk nodes of the H-shaped clock tree. At present, there is no method to automatically select the coordinates of the trunk nodes, which makes the selection of the trunk node coordinates inefficient. Summary of the invention
[0003] The technical problem to be solved by the present invention is as follows: Aiming at the problem of constructing an H-shaped clock tree in the clock tree design stage of physical design of an integrated circuit, a clustering-based H-shaped clock tree trunk node coordinate selection method and system are provided. The present invention adopts a clustering method to automatically select the trunk node coordinates of the H-shaped clock tree, avoiding manual specification of the trunk node coordinates of the H-shaped clock tree, and can accelerate the construction of the H-shaped clock tree. It has the advantages of high automation, flexibility and good adaptability, and can quickly and efficiently help construct an H-shaped clock tree in a given area.
[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0005] A clustering-based H-shaped clock tree trunk node coordinate selection method, comprising:
[0006] 1) Obtain all flip-flops in a given area in the layout-completed design, set the number of trunk nodes N of the H-shaped clock tree and initialize the coordinates of each trunk node so that each trunk node corresponds to a cluster;
[0007] 2) Assign each trigger to a cluster corresponding to a backbone node and calculate the cost value;
[0008] 3) Calculate the center point coordinates of the cluster corresponding to each backbone node;
[0009] 4) Determine whether the preset end condition is met. If the preset end condition is met, the center point coordinates of the cluster corresponding to each backbone node are used as the backbone node coordinates of the selected H-shaped clock tree, and the process ends and exits; otherwise, jump to step 2).
[0010] Optionally, the given area in step 1) is a rectangular area whose length is represented by l and whose width is represented by h.
[0011] Optionally, in step 1), setting the number N of trunk nodes of the H-shaped clock tree and initializing the coordinates of the N trunk nodes includes:
[0012] 1.1) Use vector (C1, C2, ..., C N ) represents N H-shaped clock tree trunk nodes, and the coordinates of each trunk node are expressed as (μ x (C k ),μ y (C k )) represents, where k∈[1,N]; the given area is determined as a rectangular area with m backbone nodes in the x direction and n backbone nodes in the y direction, and N=m*n, the backbone node C k The serial number k is represented by (i-1)*m+j, where i∈[1,m],j∈[1,n];
[0013] 1.2) Divide the length and width of the given area into N+1 equal parts, each part is l / (N+1) long and h / (N+1) wide;
[0014] 1.3) Through μ x (C k )=l / (N+1)*j,μ y (C k )=h / (N+1)*i Initialize the coordinates of any k-th backbone node (μ x (C k ),μ y (C k )).
[0015] Optionally, in step 2), allocating each trigger to a cluster corresponding to a backbone node and calculating the cost value means finding the backbone node closest to each trigger in geometric distance, and taking the closest geometric distance as the cost value of the trigger.
[0016] Optionally, the geometric distance is Manhattan distance, and the calculation function expression of the cost value is:
[0017] cost=min k∈[1,N] |x-μ x (C k )|+|y-μy (C k )|,
[0018] In the above formula, cost represents the cost value, min represents the minimum value in the range of k∈[1,N], k is the serial number of the backbone node, (x, y) is the coordinate of the trigger, (μ x (C k ),μ y (C k )) is any k-th backbone node C k The coordinates of .
[0019] Optionally, the function expression for calculating the center point coordinates of the cluster corresponding to each backbone node in step 3) is:
[0020]
[0021] In the above formula, (μ x (C k ),μ y (C k )) is any k-th backbone node C k The coordinates of (x i ,y i ) are the coordinates of any i-th trigger, and M is the number of triggers in the k-th cluster.
[0022] Optionally, the end condition preset in step 4) means that the cost value of each trigger no longer changes.
[0023] Optionally, the end condition preset in step 4) refers to the number of iterations of step 2) and step 3) exceeding a set value.
[0024] In addition, the present invention also provides a clustering-based H-shaped clock tree trunk node coordinate selection system, including an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the steps of the clustering-based H-shaped clock tree trunk node coordinate selection method.
[0025] In addition, the present invention also provides a computer-readable storage medium, in which a computer program is stored, and the computer program is used to be programmed or configured by a microprocessor to execute the steps of the clustering-based H-shaped clock tree trunk node coordinate selection method.
[0026] Compared with the prior art, the present invention has the following advantages:
[0027] 1. The present invention does not require manual specification of the trunk node coordinates of the H-shaped clock tree, which can effectively avoid the arbitrariness of manual specification.
[0028] 2. The present invention automatically obtains the coordinates of the trunk nodes of the H-shaped clock tree according to the specific distribution of the triggers after layout, and can quickly and efficiently generate the H-shaped clock tree. It has the advantages of high automation, high flexibility and good adaptability, and provides an efficient method for quickly constructing the H-shaped clock tree of integrated circuits. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 The figure is a flow chart of a method according to a first embodiment of the present invention.
[0030] Figure 2 Schematic diagram of an H-shaped clock tree with four trunk nodes in Embodiment 1 of the present invention.
[0031] Figure 3 Schematic diagram of initialization coordinates of a rectangular area with four trunk nodes in Embodiment 1 of the present invention.
[0032] Figure 4 This is a schematic diagram of allocating four triggers to four backbone node clusters in Embodiment 1 of the present invention. DETAILED DESCRIPTION
[0033] Embodiment 1:
[0034] like Figure 1 As shown, the clustering-based H-shaped clock tree trunk node coordinate selection method of this embodiment includes:
[0035] 1) Obtain all flip-flops in a given area in the layout-completed design, set the number of trunk nodes N of the H-shaped clock tree and initialize the coordinates of each trunk node so that each trunk node corresponds to a cluster;
[0036] 2) Traverse all triggers and take out a trigger as the current trigger, assign the current trigger to a cluster corresponding to a backbone node and calculate the cost value;
[0037] 3) Calculate the center point coordinates of the cluster corresponding to each backbone node;
[0038] 4) Determine whether all triggers have been traversed. If so, take the center point coordinates of the cluster corresponding to each trunk node as the trunk node coordinates of the selected H-shaped clock tree, end and exit; otherwise, jump to step 2).
[0039] In this embodiment, the given area in step 1) is a rectangular area whose length is represented by l and whose width is represented by h. The rectangular area can be either a rectangular area or a square area.
[0040] In this embodiment, setting the number N of trunk nodes of the H-shaped clock tree and initializing the coordinates of the N trunk nodes in step 1) includes:
[0041] 1.1) Use vector (C1, C2, ..., C N ) represents N H-shaped clock tree trunk nodes, and the coordinates of each trunk node are expressed as (μ x (C k ),μ y (C k )) represents, where k∈[1,N]; the given area is determined as a rectangular area with m backbone nodes in the x direction and n backbone nodes in the y direction, and N=m*n, the backbone node C k The serial number k is represented by (i-1)*m+j, where i∈[1,m],j∈[1,n];
[0042] 1.2) Divide the length and width of the given area into N+1 equal parts, each part is l / (N+1) long and h / (N+1) wide;
[0043] 1.3) Through μ x (C k )=l / (N+1)*j,μ y (C k )=h / (N+1)*i Initialize the coordinates of any k-th backbone node (μ x (C k ),μ y (C k )). In this embodiment, the value of the number of trunk nodes N of the H-shaped clock tree is set to 4. The schematic diagram of the H-shaped clock tree with 4 trunk nodes is as follows: Figure 2 As shown, the coordinates of the four trunk nodes are initialized as Figure 3 shown.
[0044] In this embodiment, assigning each trigger to a cluster corresponding to a backbone node and calculating the cost value in step 2) means finding the backbone node closest to each trigger in geometric distance, and taking the closest geometric distance as the cost value of the trigger.
[0045] The geometric distance can be calculated in the desired distance calculation method according to the needs. For example, as an optional implementation, the geometric distance in this embodiment is the Manhattan distance, and the calculation function expression of the cost value is:
[0046] cost=min k∈[1,N] |x-μ x (C k )|+|y-μ y (C k )|,
[0047] In the above formula, cost represents the cost value, min represents the minimum value in the range of k∈[1,N], k is the serial number of the backbone node, (x, y) is the coordinate of the trigger, (μ x(C k ),μ y (C k )) is any k-th backbone node C k The coordinates of .
[0048] In this embodiment, the function expression for calculating the center point coordinates of each cluster corresponding to each backbone node in step 3) is:
[0049]
[0050] In the above formula, (μ x (C k ),μ y (C k )) is any k-th backbone node C k The coordinates of (x i ,y i ) is the coordinate of any i-th trigger, and M is the number of triggers in the k-th cluster. Figure 4 As shown, the four triggers FF1, FF2, FF3 and FF4 in a given area are respectively assigned to the clusters corresponding to the backbone nodes C1, C2, C3 and C4 according to the Manhattan distance.
[0051] The end condition preset in step 4) can be selected as needed. For example, as an optional implementation, the end condition preset in step 4) of this embodiment refers to that the cost value of each trigger no longer changes. If it is found that the cost value of each trigger is still changing, that is, it is different from the result obtained by executing step 2) last time, then jump to repeat step 2)-step 3) until the cost value of all triggers no longer changes, and the final center point coordinates of each cluster are used as the coordinates of the trunk node of the H-shaped clock tree.
[0052] In summary, the aforementioned clustering-based H-shaped clock tree trunk node coordinate selection method of this embodiment constructs an H-shaped clock tree in a rectangular area after layout, and automatically obtains the trunk node coordinates of the H-shaped clock tree by a clustering method, so that the H-shaped clock tree can be constructed quickly and efficiently, avoiding manual specification of the trunk node coordinates of the H-shaped clock tree. It has the advantages of high automation, flexibility and good adaptability, and can help to quickly and efficiently construct the H-shaped clock tree in a given area.
[0053] In addition, this embodiment also provides a clustering-based H-shaped clock tree trunk node coordinate selection system, including an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the steps of the aforementioned clustering-based H-shaped clock tree trunk node coordinate selection method.
[0054] In addition, this embodiment also provides a computer-readable storage medium, in which a computer program is stored, and the computer program is used to be programmed or configured by a microprocessor to execute the steps of the aforementioned clustering-based H-shaped clock tree trunk node coordinate selection method.
[0055] Embodiment 2:
[0056] This embodiment is basically the same as the first embodiment, and the main difference is that the end condition preset in step 4) is different. In this embodiment, the end condition preset in step 4) refers to the number of iterations of step 2) and step 3) exceeding the set value. By taking the value of the set value, the effect similar to that of the first embodiment can be achieved.
[0057] In addition, this embodiment also provides a clustering-based H-shaped clock tree trunk node coordinate selection system, including an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the steps of the aforementioned clustering-based H-shaped clock tree trunk node coordinate selection method.
[0058] In addition, this embodiment also provides a computer-readable storage medium, in which a computer program is stored, and the computer program is used to be programmed or configured by a microprocessor to execute the steps of the aforementioned clustering-based H-shaped clock tree trunk node coordinate selection method.
[0059] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the functions in the process. Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0060] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.
Claims
1. A clustering-based H-shaped clock tree trunk node coordinate selection method, characterized in that: include: 1) Get all the flip-flops in the given area of the layout-completed design and set the number of trunk nodes in the H-shaped clock tree N And initialize the coordinates of each backbone node so that each backbone node corresponds to a cluster; 2) Assign each trigger to a cluster corresponding to a backbone node and calculate the cost value; 3) Calculate the center point coordinates of the cluster corresponding to each backbone node; 4) Determine whether the preset end condition is met. If the preset end condition is met, the center point coordinates of the cluster corresponding to each backbone node are used as the backbone node coordinates of the selected H-shaped clock tree, and the process ends and exits. Otherwise, jump to step 2); The given area in step 1) is long l Indicates, wide use h Represents the rectangular area; In step 1), setting the number of trunk nodes N of the H-shaped clock tree and initializing the coordinates of the N trunk nodes includes: 1.1) Use vector ( C 1, C 2, …, C N )express N H-shaped clock tree trunk nodes, the coordinates of each trunk node are ( μ x ( C k ), μ y ( C k )) indicates that k ∈[1, N ]; Determine the rectangular area within the given area x Direction m backbone nodes, y Direction n backbone nodes, and satisfy N = m * n , backbone node C k Serial number k Expressed as ( i -1)* m + j ,in i ∈[1, m ], j ∈[1, n ]; 1.2) Divide the length and width of the given area into N +1 equal parts, each part is l / ( N +1), with a width of h / ( N +1); 1.3) By μ x ( C k ) = l / ( N +1)* j , μ y ( C k ) = h / ( N +1)* i Initialize any k The coordinates of the backbone nodes ( μ x ( C k ), μ y ( C k )).
2. The clustering-based H-shaped clock tree trunk node coordinate selection method according to claim 1, characterized in that: In step 2), allocating each trigger to a cluster corresponding to a backbone node and calculating the cost value means finding the backbone node with the closest geometric distance to each trigger, and taking the closest geometric distance as the cost value of the trigger.
3. The clustering-based H-shaped clock tree trunk node coordinate selection method according to claim 2, characterized in that: The geometric distance is Manhattan distance, and the calculation function expression of the cost value is: , In the above formula, cost represents the cost value, min represents k ∈[1, N ] take the minimum value in the range, k is the serial number of the main node, (x, y) is the coordinate of the trigger, ( μ x ( C k ), μ y ( C k )) is any k Backbone Node C k The coordinates of .
4. The clustering-based H-shaped clock tree trunk node coordinate selection method according to claim 1, characterized in that: The function expression for calculating the center point coordinates of the cluster corresponding to each backbone node in step 3) is: , In the above formula, ( μ x ( C k ), μ y ( C k )) is any k Backbone Node C k The coordinates of ( x i , y i ) is the coordinate of any i-th trigger, M For the k The number of triggers in a cluster.
5. The clustering-based H-shaped clock tree trunk node coordinate selection method according to claim 1, characterized in that: The end condition preset in step 4) means that the cost value of each trigger no longer changes.
6. The clustering-based H-shaped clock tree trunk node coordinate selection method according to claim 1, characterized in that: The preset end condition in step 4) refers to the number of iterations of step 2) and step 3) exceeding the set value.
7. A clustering-based H-shaped clock tree trunk node coordinate selection system, comprising a microprocessor and a memory connected to each other, characterized in that: The microprocessor is programmed or configured to execute the steps of the clustering-based H-shaped clock tree trunk node coordinate selection method as described in any one of claims 1 to 6.
8. A computer-readable storage medium having a computer program stored therein, characterized in that: The computer program is used to be programmed or configured by a microprocessor to execute the steps of the clustering-based H-shaped clock tree trunk node coordinate selection method as described in any one of claims 1 to 6.
