A standard cell layout-oriented drc automatic repair strategy

Abstract

The application discloses a DRC automatic repair strategy for standard cell layout, and relates to electric digital data processing.The strategy is as follows: identifying DRC violation in the standard cell layout; extracting graphic information of GDS file of the standard cell layout and taking the graphic information as current solution x; dynamically adjusting temperature attenuation coefficient of simulated annealing algorithm according to the current solution x and iteration number to optimize the current solution x, and taking the optimal solution obtained by optimization as DRC violation repair parameter of the standard cell layout to repair the DRC violation.The application has obvious advantages in efficiency, accuracy and stability, and can effectively solve the deficiency of traditional artificial repair mode, and provides strong support for integrated circuit layout design.

Description

TECHNICAL FIELD

[0001] The present application relates to electric digital data processing, more particularly, it relates to a DRC automatic repair strategy for standard cell layout. BACKGROUND

[0002] The standard cell library is an important bridge of digital integrated circuits, and the layout design in its construction process is the most complex and time-consuming, which is mainly due to the fact that the layout design needs to follow extremely complex design rules, which are mutually restrictive and interlocking. At present, the existing technical solutions mostly adopt manual methods when trying to solve the DRC violation problem of standard cell layout, mainly relying on engineers to check the violation points in the layout one by one and manually adjust, which is not only low in efficiency, but also easily disturbed by human factors, resulting in uneven repair quality. SUMMARY

[0003] The technical problem to be solved by the present application is to overcome the deficiencies of the prior art, and to provide a DRC automatic repair strategy for standard cell layout, aiming to effectively overcome the many difficulties in the DRC violation of standard cell layout design, and to realize efficient and accurate automatic repair of DRC violation of standard cell layout by innovatively using improved simulated annealing algorithm. The strategy can intelligently identify various complex violation conditions, and fully consider the overall structure of the layout and various design rules during the repair process, ensuring that the DRC violation is eliminated while the original performance and function of the layout are maintained to the greatest extent, providing a more reliable and efficient solution for standard cell layout design.

[0004] The DRC automatic repair strategy for standard cell layout provided by the present application is as follows: identifying DRC violations in the standard cell layout; extracting the graphic information of the GDS file of the standard cell layout, and taking the image information as the current solution x; dynamically adjusting the temperature decay coefficient of the simulated annealing algorithm according to the current solution x and the number of iterations to optimize the current solution x, and taking the optimal solution obtained by optimization as the DRC violation repair parameter of the standard cell layout to repair the DRC violation.

[0005] Preferably, the specific steps of dynamically adjusting the temperature decay coefficient of the simulated annealing algorithm are as follows:

[0006] S101, initializing the initial temperature T0, the termination temperature T end , the initial temperature decay coefficient a0, and the maximum number of iterations iter max of the simulated annealing algorithm.

[0007] S102, randomly generating an initial solution x0.

[0008] S103, calculate the objective function value f(x0) of the initial solution x0;

[0009] S104, set the initial solution x0 as the historical optimal solution x best , and initialize the objective function of the historical optimal solution x best ;

[0010] S105, judge whether the current temperature T is greater than the termination temperature T end and the current iteration number iter is less than the maximum iteration number iter max , if yes, proceed to S106; otherwise, execute the DRC violation repair judgment;

[0011] S106, calculate a new temperature attenuation coefficient a according to the initial temperature attenuation coefficient a0, the current iteration number iter, the maximum iteration number iter max , and the objective function of the historical optimal solution x best ; new ; at the same time, update the current temperature T to a new ×T;

[0012] S107, heuristically generate a field solution x';

[0013] S108, calculate the objective function value f(x') of the field solution x' and the objective function value f(x) of the current solution x;

[0014] S109, determine whether to accept the field solution x' as a new current solution according to the objective function value f(x') of the field solution x' and the objective function value f(x) of the current solution x, if yes, execute S110;

[0015] S110, judge whether the objective function value f(x') of the field solution x' is less than the objective function value f(x best ) of the historical optimal solution x best , if yes, execute S111; otherwise, execute S112;

[0016] S111, update the historical optimal solution x best and the objective function value f(x best ) thereof to the field solution x' and the objective function value f(x') thereof;

[0017] S112, assign the minimum value of the objective function value of the historical optimal solution x best to f min = min(f min ,f(x')) and the maximum value to f max = max(f max ,f(x'));

[0018] S113, calculate a new solution acceptance probability P of the new current solution, judge whether the new current solution is better than the historical optimal solution x best and whether the new solution acceptance probability P is less than a set minimum probability threshold; if the new current solution is better than the historical optimal solution x best and the new solution acceptance probability P is less than the minimum probability threshold, restore the historical optimal solution x best as the current solution and as the optimal solution, then update the current iteration number iter, and return to S105; otherwise, take the field solution x' as the optimal solution, update the current iteration number iter, and return to S105.

[0019] Preferably, in S105, the DRC violation repair judgment is specifically:

[0020] check the repair result of the DRC violation, if the DRC violation still exists, execute S101, if the DRC violation does not exist, end the repair of the DRC violation.

[0021] Preferably, in S104, the historical optimal solution x best is initialized as:

[0022] the target function value f(x0) of the calculated initial solution x0 is assigned to the target function value f(x best ) of the historical optimal solution x best , the minimum value f best and the maximum value f best of the target function value f(x min ) of the historical optimal solution x max .

[0023] Preferably, in S106, the new temperature attenuation coefficient a new is calculated by:

[0024]

[0025] wherein f(x) = f(x0); β and λ are both adjustment parameters.

[0026] Preferably, in S107, the method for generating the field solution x' by heuristic is:

[0027] calculate the conflict number between each pattern and other patterns in the standard cell layout, sort the patterns according to the conflict number from large to small, select the pattern with the largest conflict number for position adjustment, and generate the field solution x'.

[0028] Preferably, the calculation method of the conflict number is:

[0029] Obtaining the distance d between two patterns ij Setting a minimum distance constraint d min If d ij d min If d

[0030] Preferably, in S108, the objective function value f(x) of the current solution x is calculated by the following formula:

[0031]

[0032] Where n is the number of patterns in the standard cell layout, d ij is the distance between pattern i and pattern j.

[0033] Preferably, in S109, the specific method for determining whether to accept the field solution x' as the new current solution is as follows:

[0034] If f(x') < f(x), the field solution x' is accepted as the new current solution.

[0035] If f(x') ≥ f(x), a new solution acceptance probability P is calculated, and a random number γ is generated; if γ < P, the field solution x' is accepted as the new current solution; otherwise, the current solution x remains unchanged.

[0036] Preferably, the calculation formula of the new solution acceptance probability P is as follows:

[0037]

[0038] Advantages

[0039] The present application has the following advantages:

[0040] 1. The repair efficiency is greatly improved:

[0041] The present application adopts an automatic repair strategy and uses an improved simulated annealing algorithm to quickly identify and process violation problems. The algorithm efficiently searches for the optimal solution in the solution space through an adaptive temperature decay strategy and a heuristic neighborhood solution generation method, greatly shortens the repair time, and significantly improves the repair efficiency.

[0042] 2. The repair accuracy is higher:

[0043] The application can automatically identify various complex violation cases, ensure the accuracy of the repair process based on accurate objective function calculation and strict solution acceptance criteria. For example, by calculating the number of conflicts between graphics to adjust the graphics with serious conflicts first, key problems can be solved more targetedly, negative effects on other parts of the layout during the repair process can be effectively avoided, thereby ensuring that the repaired layout is more in line with the design rule requirements and improving the accuracy of repair.

[0044] 3、Reduce human factor interference:

[0045] The automatic repair strategy of the application is not disturbed by human emotions, fatigue and other factors. As long as the algorithm parameters are set reasonably, it can operate stably according to the established rules and processes, and each repair follows the same scientific method, thereby ensuring the stability and reliability of the repair results and providing more reliable protection for standard cell layout design. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 The DRC automatic repair strategy flowchart of the application;

[0047] Figure 2 The improved simulated annealing algorithm flowchart of the application;

[0048] Figure 3 The standard cell layout DRC violation repair schematic diagram of the application. DETAILED DESCRIPTION

[0049] The application will be further described below in conjunction with the embodiments, but does not constitute any limitation on the application. Any limited modification made by anyone within the scope of the claims of the application is still within the scope of the claims of the application.

[0050] Referring to Figure 1 The application is a standard cell layout-oriented DRC automatic repair strategy, which mainly includes the following three steps.

[0051] S1, identify the DRC violation in the standard cell layout.

[0052] In standard cell layout design, design rule check (DRC) is a key step to ensure that the layout meets the requirements of the manufacturing process. Through professional automated tools (such as Calibre), a comprehensive DRC violation check is performed on the standard cell layout. These tools will check each graphic element in the layout according to a series of pre-set design rules (such as line width, spacing, coverage, etc.). After identifying all types of violations, they are classified. Common violation types include: (1) spacing violation: the distance between graphics is less than the minimum spacing specified by the design rules. (2) line width violation: the width of the conductor is less than or greater than the range allowed by the design rules. (3) area violation: the area of the graphic does not meet the requirements of the design rules. For example Figure 3 As shown in the figure, (a) contact hole and metal spacing; (b) metal and metal spacing; (c) contact hole and active area spacing.

[0053] S2, extract the graphic information of the GDS file of the standard cell layout, and take the image information as the current solution x. Specifically, the GDS file of the standard cell layout is consulted using Klayout software, and the graphic information is extracted to form specific data point coordinates. That is, the data point coordinates are taken as the current solution x to meet the requirements of the input parameters in the simulated annealing algorithm.

[0054] S3, dynamically adjust the temperature decay coefficient of the simulated annealing algorithm according to the current solution x and the number of iterations, to optimize the current solution, and take the optimal solution as the DRC violation repair parameter of the standard cell layout to repair the DRC violation.

[0055] The simulated annealing algorithm simulates the solid annealing process, searches randomly in the solution space, and accepts inferior solutions with a certain probability, so as to avoid falling into local optimal solution, and finally find the global optimal solution. In the repair of standard cell layout, the goal is to minimize the mutual conflict between graphics, and ensure that the distance between all graphics meets the minimum distance constraint defined by the user.

[0056] In the search process of the traditional simulated annealing algorithm, the temperature is linearly decreased according to the fixed decay coefficient, and the generation of neighborhood solution is completely random. In the problem of repairing standard cell layout, this fixed temperature decay and random neighborhood solution generation method may lead to slow convergence speed of the algorithm, or the risk of falling into local optimal solution in the later search stage is still high. Therefore, the present application proposes an improved simulated annealing algorithm, which adopts an adaptive temperature decay strategy, dynamically adjusts the temperature decay coefficient according to the current solution x and the number of iterations of the search, so as to better repair the DRC error of the standard cell.

[0057] For the improved simulated annealing algorithm, as Figure 2 shown, the specific implementation steps are as follows:

[0058] S101, initialize initial temperature T0, termination temperature T, initial temperature decay coefficient a0, maximum iteration number iter of the simulated annealing algorithm end max For example, T0 is set to 100, T end is set to 0.1, a0 is set to 0.8-0.99, iter max is set to 100-1000, and the iteration number L at each temperature is set to 10-100.

[0059] The following is an explanation of important parameters:

[0060] Initial temperature T0: The initial temperature is an important parameter of the simulated annealing algorithm, which determines the randomness of the algorithm in the early search. A higher initial temperature allows the algorithm to search more widely in the solution space, avoiding premature convergence to a local optimal solution. Generally, a suitable initial temperature can be selected according to the size and complexity of the problem, for example, for a small-scale layout repair problem, set T0 to 100 first.

[0061] Termination temperature T end : The termination temperature represents the condition for the algorithm to stop searching. When the temperature drops below the termination temperature, the algorithm considers that a better solution has been found and stops iteration. Usually, T end is set to a small value, such as 0.1.

[0062] Temperature decay coefficient a: The temperature decay coefficient controls the speed of temperature drop. Its value is usually between 0 and 1, for example, a = 0.9 means that the temperature is reduced to 90% of the original after each iteration.

[0063] Iteration number L at each temperature: At each temperature, the algorithm will perform L iterations to fully explore the solution space at the current temperature. The value of L can be adjusted according to the complexity of the problem, and it can be set to between 10 and 100.

[0064] S102, randomly generate an initial solution x0. The initial solution x0 represents a layout of the graphics in the standard cell layout. The initial solution can be generated by randomly moving or rotating the graphics.

[0065] S103, calculate the objective function value f(x0) of the initial solution x0.

[0066] S104, set the initial solution x0 as the historical optimal solution x best , and initialize the objective function of the historical optimal solution x best . In this step, the initialization of the objective function of the historical optimal solution x best includes:

[0067] ​The objective function value f(x0) of the calculated initial solution x0 is assigned to the historical optimal solution x best at the same time best , the minimum value f best and the maximum value f best of the objective function value f(x min ) of the historical optimal solution x max ). That is: x best =x0, f(x best )=f(x0), f min =f(x0), f max =f(x0).

[0068] S105, judge whether the current temperature T is greater than the termination temperature T end and the current iteration number iter is less than the maximum iteration number iter max , if yes, proceed to S106; otherwise, perform DRC violation repair judgment to determine whether the DRC violation is completely repaired.

[0069] S106, adaptive temperature attenuation: calculate a new temperature attenuation coefficient a new according to the initial temperature attenuation coefficient a0, the current iteration number iter, the maximum iteration number iter best , and the objective function of the historical optimal solution x max .

[0070] In this embodiment, the new temperature attenuation coefficient a new is calculated by the following formula:

[0071]

[0072] In the formula, f(x)=f(x0); β and λ are both adjustment parameters, and their values are between 0.5 and 2.

[0073] Then update the current temperature T to a new ×T, that is, T=a new T.

[0074] S107, generate a heuristic field solution x'. The generation method of the field solution x' is as follows:

[0075] First, calculate the conflict times between each pattern in the standard cell layout and other patterns. For two patterns i and j, get the distance d ij between them, if d ij <d min (d minIf the distance between the two patterns is less than the minimum distance constraint defined by the user (i.e. d(i, j) < dmin), it is considered that the two patterns are in conflict, and the conflict number of the patterns i and j is added by 1 respectively. Then the patterns are sorted according to the conflict number from large to small. In the generation of the neighborhood solution, the pattern with the largest conflict number is selected to adjust the position to generate the neighborhood solution x'. For example, a random small offset between [-0.1, 0.1] is added to the x and y coordinates of the pattern with the largest conflict number to adjust the position.

[0076] S108, the objective function value f(x') of the neighborhood solution x' and the objective function value f(x) of the current solution x are calculated. In the embodiment, the objective function value f(x) is defined as the conflict number between the patterns that do not satisfy the minimum distance constraint:

[0077]

[0078] In the formula, n is the number of patterns in the standard cell layout, d ij is the distance between the pattern i and the pattern j.

[0079] S109, solution acceptance: whether to accept the neighborhood solution x' as the new current solution is determined according to the objective function value f(x') of the neighborhood solution x' and the objective function value f(x) of the current solution x. If accepted, S110 is executed.

[0080] Specifically, the specific method for determining whether to accept the neighborhood solution x' as the new current solution is as follows:

[0081] If f(x') < f(x), it means that the new solution is better than the current solution, and the neighborhood solution x' is accepted as the new current solution, that is, the current solution x is updated as x = x'.

[0082] If f(x') ≥ f(x), the acceptance probability of the new solution is calculated A random number γ is generated at the same time; wherein γ is any random number between 0 and 1. If γ < P, the neighborhood solution x' is accepted as the new current solution, that is, the current solution x is updated as x = x'. Otherwise, the current solution x remains unchanged.

[0083] S110, whether the objective function value f(x') of the neighborhood solution x' is less than the objective function value f(x best ) of the historical optimal solution x best is determined. If yes, it means that the new solution is the historically optimal solution, and S111 is executed. Otherwise, S112 is executed.

[0084] S111, the historical optimal solution x best and the objective function value f(x best ) thereof are updated as the neighborhood solution x' and the objective function value f(x') thereof, that is, x best = x', f(x best) = f(x').

[0085] S112, assign the minimum value of the objective function value of the historical optimal solution x best to f min = min(f min , f(x')) and the maximum value to f max = max(f max , f(x')).

[0086] After the assignment of the current solution x in the above steps, we get a new current solution, which can be the original current solution x or the domain solution x', but whether the new current solution is the optimal solution needs to be further compared and determined.

[0087] S113, first, we need to determine whether the objective function value of the new current solution is worse than the objective function value f(x best ) of the historical optimal solution x best , i.e., if f(x) ≥ f(x best ), it means that the new current solution is worse than the historical optimal solution x best . Then we need to determine the relationship between the new solution acceptance probability P and the set minimum probability threshold. If f(x) ≥ f(x best ) and the new solution acceptance probability P is less than the set minimum probability threshold, the historical optimal solution x best is restored as the current solution, and the current solution with the restored historical optimal solution x best is taken as the optimal solution, then the current iteration number iter is updated, i.e., iter = iter + 1, and returns to S105. If either of the two judgment conditions f(x) ≥ f(x end ) and the new solution acceptance probability P is less than the set minimum probability threshold is not true, the domain solution x' is taken as the optimal solution, the current iteration number iter is updated, i.e., iter = iter + 1, and returns to S105. After each iteration, the randomness of the algorithm also gradually decreases, so that the algorithm gradually converges to a better solution.

[0088] In the improved simulated annealing algorithm described above, the termination condition is: when T ≤ T max or iter ≥ iter end , the algorithm terminates. That is, in the above S105, if the previous temperature T is not greater than the termination temperature T max and the current iteration number iter is not less than the maximum iteration number iter 2 , the algorithm terminates.If yes, the algorithm terminates, and then a DRC violation repair judgment is performed. The specific content of the judgment is: checking the repair result of the DRC violation, if the DRC violation still exists, performing S101, if the DRC violation does not exist, ending the repair of the DRC violation. In actual application, a maximum iteration number can be set to avoid the algorithm from falling into an infinite loop.

[0089] In the embodiment, the time complexity of the improved simulated annealing algorithm mainly depends on the iteration number and the calculation complexity of the objective function. In each iteration, the value of the objective function and the conflict number of each graph need to be calculated, and the calculation of the objective function needs to traverse all graph pairs, and the time complexity is O(n 2 ), wherein n is the number of graphs. The total iteration number depends on the initial temperature, the termination temperature and the maximum iteration number, and is generally O(iter max ). Therefore, the total time complexity of the improved simulated annealing algorithm is O(n 2 ·iter max ). In terms of space complexity, the main space overhead is in storing the graph information, the conflict number and the historical optimal solution, and therefore the space complexity is O(n), wherein n is the number of graphs. It can be seen that the method of the present application for dynamically adjusting the temperature decay coefficient based on the current solution and the search iteration number is the key to distinguishing from the traditional algorithm and improving the DRC repair effect of the layout; and generating the neighborhood solution by preferentially adjusting the position of the graph with the conflict number helps to improve the efficiency and accuracy of the repair algorithm.

[0090] The above only describes the preferred embodiments of the present application, and it should be noted that for those skilled in the art, without departing from the structure of the present application, a number of modifications and improvements can be made, which will not affect the effect of the present application and the practicality of the patent.

Claims

1. A DRC automatic repair strategy for standard cell layout, characterized in that, The strategy is: identifying DRC violation in standard cell layout; extracting graphic information of GDS file of the standard cell layout, and taking the graphic information as current solution x ; According to the current solution x And the number of iterations dynamic adjustment of simulated annealing algorithm temperature attenuation coefficient, to the current solution x Optimization, and the optimal solution obtained by optimization as the DRC violation repair parameters of the standard cell layout DRC violation repair; The specific steps of dynamically adjusting the temperature attenuation coefficient of the simulated annealing algorithm are as follows: S101、initialize an initial temperature of the simulated annealing algorithm , a termination temperature , an initial temperature attenuation coefficient , a maximum iteration number ; S102、Randomly generate an initial solution x 0; S103、compute the initial solution x 0 of the objective function value ; S104, set the initial solution x 0 is the historical best solution and initialize the historical best solution of the objective function; S105, judging whether the current temperature T is greater than the termination temperature and the current iteration number is less than the maximum iteration number , if yes, proceeding to S106; otherwise, performing DRC violation repair judgment; S106、calculating a new temperature decay coefficient according to the initial temperature decay coefficient , the current iteration number , the maximum iteration number , the historical optimal solution of the objective function ; meanwhile, updating the current temperature T to ×T; S107, generating the domain solution by heuristic x’ ; S108, compute the objective function value of the field solution x’ x ;​​​ S109, determine whether to accept the field solution as a new current solution based on the objective function value of the field solution and the objective function value of the current solution x’ x x’ S110, if the field solution is accepted as a new current solution, execute S111​​​​ S110, Determine the domain solution x’ objective function value Is it less than the historical best solution? objective function value If yes, then execute S111; otherwise, execute S112. S111, updating the history optimal solution and its objective function value to the domain solution x’ and its objective function value ; S112, assign the minimum value of the objective function value of the historical optimal solution to the minimum value of the objective function value of the historical optimal solution to the maximum value of the objective function value of the historical optimal solution ; S113, calculate a new solution acceptance probability P of the new current solution, judge whether the new current solution is worse than the historical optimal solution and whether the new solution acceptance probability P is less than a set minimum probability threshold; if the new current solution is worse than the historical optimal solution and the new solution acceptance probability P is less than the minimum probability threshold, restore the historical optimal solution as the current solution and as the optimal solution, then update the current iteration number and return to S105; otherwise, take the domain solution x’ as the optimal solution, update the current iteration number and return to S105.

2. The DRC automatic repair strategy for standard cell layout according to claim 1, wherein, In S105, the DRC violation repair judgment is specifically as follows: The repair result of the DRC violation is checked, if the DRC violation still exists, S101 is executed, and if the DRC violation does not exist, the DRC violation repair is ended.

3. The DRC automatic repair strategy for standard cell layout according to claim 1, wherein, In S104, the historical optimal solution is initialized The objective function includes: the initial solution x the objective function value of the initial solution is simultaneously assigned to the historical best solution the objective function value of the historical best solution the historical best solution the objective function value of the historical best solution the minimum value and the maximum value .

4. The DRC automatic repair strategy for standard cell layout according to claim 3, wherein, In S106, the new temperature decay coefficient is calculated by the following equation: ； wherein ; β and Lambda are adjustment parameters.

5. The DRC automatic repair strategy for standard cell layout according to claim 1, wherein, In S107, the domain solution is generated by heuristics x’ The method is as follows: counting the number of conflicts between each pattern and other patterns in the standard cell layout, sorting the patterns according to the number of conflicts from large to small, selecting a pattern with the largest number of conflicts for position adjustment, and generating a region solution x’ .

6. The DRC automatic repair strategy for standard cell layout according to claim 5, wherein, The calculation method of the conflict number is as follows: Obtaining the distance between two figures , setting a minimum distance constraint , if , judging that there is a conflict between the two figures, and adding 1 to the conflict times of the two figures.

7. The DRC automatic repair strategy for standard cell layout according to claim 6, wherein, In S108, the current solution x the objective function value is calculated by the following equation: ； wherein n is the number of patterns in the standard cell layout, d ij is the distance between patterns i and patterns j .

8. The DRC automatic repair strategy for standard cell layout according to claim 1, wherein, In S109, it is judged whether to accept the domain solution x’ A specific method for the new current solution is: If then accept the field solution x’ as the new current solution; If , then compute a new solution acceptance probability P and generate a random number ; if , then accept the domain solution x’ as the new current solution; otherwise, the current solution x remains unchanged.

9. The DRC automatic repair strategy for standard cell layout according to claim 1 or 8, wherein, The calculation formula of the new solution acceptance probability P is as follows: 。

Images