CN110717310A

CN110717310A - Method and device for adjusting circuit layout

Info

Publication number: CN110717310A
Application number: CN201910959759.5A
Authority: CN
Inventors: 吴玉平; 陈岚; 张学连
Original assignee: Institute of Microelectronics of CAS
Current assignee: Institute of Microelectronics of CAS
Priority date: 2019-10-10
Filing date: 2019-10-10
Publication date: 2020-01-21

Abstract

The embodiment of the application discloses a method and a device for adjusting a circuit layout, which are used for determining adjacent nets in the layout to be adjusted according to the adjacent relation of the nets in the layout to be adjusted, wherein the adjacent nets can comprise a first net and a second net, simulating a circuit netlist comprising the adjacent nets, judging whether the adjacent nets have functional destructive crosstalk according to a simulation waveform, if so, further determining the maximum crosstalk capacitance without the functional destructive crosstalk in the adjacent nets, calculating the optimized adjacent parameters of the adjacent nets according to the maximum crosstalk capacitance and the initial adjacent parameters of the adjacent nets, adjusting the adjacent nets based on the optimized adjacent parameters to ensure that the actual crosstalk capacitance in the adjacent nets with the optimized adjacent parameters is less than or equal to the maximum crosstalk capacitance, so that the functional destructive crosstalk does not exist between the adjusted adjacent nets, and automatically determining the adjacent nets to be adjusted, and determining suitable proximity parameters for the neighboring nets, thereby improving the efficiency of layout design.

Description

Method and device for adjusting circuit layout

Technical Field

The present invention relates to the field of layout design of semiconductor integrated circuits, and more particularly, to a method and apparatus for adjusting a circuit layout.

Background

With the progress of the integrated circuit manufacturing process, on one hand, the spacing between the physical patterns is continuously reduced, the density of the physical connecting lines is continuously improved, on the other hand, the integration scale is continuously improved, and the number of the long physical connecting lines is continuously increased. This makes integrated circuits susceptible to crosstalk problems, i.e., parasitic capacitances between geometrically adjacent nets, through which electrical signals on one net can couple to another net, thereby disrupting the signal integrity of the other net, e.g., crosstalk of one net to another can cause distortion of signals on the other net.

Strictly speaking, signal crosstalk between geometrically adjacent nets is unavoidable, but in order to ensure that the circuit system works properly, it is necessary to ensure that the crosstalk between nets does not affect the function of the nets, i.e., to avoid the occurrence of crosstalk in the circuit that is destructive to the function of the circuit. Currently, layout constraints related to crosstalk, such as the proximity relationship between the physical wires of nets, the spacing between the physical wires of adjacent nets, the parallel length of the physical wires of adjacent nets, etc., can be manually specified before designing a circuit layout, so as to drive the layout, thereby avoiding the occurrence of crosstalk that can damage the circuit function.

In order to find out the functional destructive crosstalk noise in the physical layout, a circuit netlist containing parasitic parameters can be extracted from complete layout data so as to perform circuit simulation, and then circuit simulation waveforms are manually analyzed to find out the line network and the physical connection line thereof related to the functional destructive crosstalk noise. This results in a large number of layout iterations to eliminate the functional destructive crosstalk noise in the physical layout process, which is time consuming and limits the improvement of layout design efficiency.

Disclosure of Invention

In order to solve the above technical problems, embodiments of the present application provide a method and an apparatus for adjusting a circuit layout, which can automatically identify a neighboring net with a functional destructive crosstalk noise, and determine a suitable layout constraint condition for the neighboring net, thereby reducing the number of iterations of the layout and improving the efficiency of the layout design.

The embodiment of the application provides a method for adjusting a circuit layout, which comprises the following steps:

according to the adjacent relation of the nets in the layout to be adjusted, determining adjacent nets in the layout to be adjusted, wherein the adjacent nets comprise a first net and a second net which are adjacent;

simulating the circuit netlist including the adjacent nets, and judging whether the adjacent nets have functional destructive crosstalk according to simulation waveforms;

if so, determining that the maximum crosstalk capacitance of the functional destructive crosstalk does not exist in the adjacent nets;

calculating an optimized proximity parameter of the adjacent net according to the maximum crosstalk capacitance and the initial proximity parameter of the adjacent net, so as to adjust the adjacent net based on the optimized proximity parameter; wherein the actual crosstalk capacitance present in the neighboring nets having the optimized neighbor parameter is less than or equal to the maximum crosstalk capacitance.

Optionally, the determining a maximum crosstalk capacitance of the adjacent nets without functional destructive crosstalk includes:

adjusting the crosstalk capacitance value of the adjacent net, and performing circuit simulation on the adjacent net;

establishing a crosstalk capacitance-line network waveform model of the first line network and a crosstalk capacitance-line network waveform model of the second line network according to the plurality of groups of crosstalk capacitance values and corresponding circuit simulation results;

calculating the maximum crosstalk capacitance which can be borne by the first wire mesh according to the crosstalk capacitance-wire mesh waveform model of the first wire mesh, and calculating the maximum crosstalk capacitance which can be borne by the second wire mesh according to the crosstalk capacitance-wire mesh waveform model of the second wire mesh;

and selecting the minimum value of the maximum crosstalk capacitance which can be borne by the first wire mesh and the maximum crosstalk capacitance which can be borne by the second wire mesh as the maximum crosstalk capacitance which does not have the functional destructive crosstalk in the adjacent wire meshes.

Optionally, the simulating the circuit netlist including the adjacent net, and determining whether the adjacent net has functional destructive crosstalk according to the simulation waveform of the adjacent net includes:

performing front simulation and back simulation on the circuit netlist including the adjacent nets to obtain front simulation waveforms and back simulation waveforms;

if the changes of the front simulation waveform and the rear simulation waveform of the first net are inconsistent in a first time period, further determining whether the front simulation waveform of the second net has a waveform change corresponding to the rear simulation waveform of the first net within a preset time before the first time period;

if so, determining whether a latch or a trigger of which the input end is connected with the first net has a change corresponding to the post-simulation waveform of the first net or not in a second time period; the second time period is a preset time period after the latest clock signal effective time point after the first time period starts;

if yes, determining that the first net in the adjacent nets has functional destructive crosstalk;

or the like, or, alternatively,

if the latch or the trigger with the input end connected with the first net generates the change inconsistent with the previous simulation waveform of the first net in a second time period, determining a clock signal time window; the second time period is a preset time period after the latest clock signal effective time point after the first time period starts;

judging that the changes of the front simulation waveform and the rear simulation waveform of the first net in the clock signal time window are not consistent, if so, further determining whether the front simulation waveform of the second net in the clock signal time window has a waveform change corresponding to the rear simulation waveform of the first net;

and if so, determining that the first net in the adjacent nets has functional destructive crosstalk.

Optionally, the initial proximity parameter includes an initial parallel line length and an initial line spacing between the first net and the second net, and the calculating, according to the maximum crosstalk capacitance and the initial proximity parameter, to obtain an optimized proximity parameter includes:

calculating to obtain a feasible connecting line distance according to the initial parallel line length and the maximum crosstalk capacitance;

if the feasible connecting line spacing is smaller than or equal to a first preset spacing, taking the first preset spacing as an optimized connecting line spacing;

if the feasible connecting line spacing is larger than the first preset spacing and is smaller than or equal to a second preset spacing, taking the feasible connecting line spacing as an optimized connecting line spacing;

and taking the initial parallel line length and the optimized connecting line distance as optimized adjacent parameters.

and calculating to obtain an optimized parallel line length according to the optimized connecting line distance and the maximum crosstalk capacitance, and taking the optimized parallel line length and the optimized connecting line distance as optimized adjacent parameters.

calculating to obtain an optimized parallel line length according to the initial connection distance and the maximum crosstalk capacitance;

and taking the initial connecting line distance and the optimized parallel line length as optimized adjacent parameters.

Optionally, the method further includes:

calculating the initial parasitic capacitance of the adjacent net according to the initial adjacent parameters;

if the initial parasitic capacitance is smaller than the maximum crosstalk capacitance and the difference between the maximum crosstalk capacitance and the initial parasitic capacitance is larger than a preset value, calculating an optimized adjacent parameter of the adjacent net according to the maximum crosstalk capacitance and the initial adjacent parameter of the adjacent net so as to adjust the adjacent net based on the optimized adjacent parameter; and the actual crosstalk capacitance existing in the adjacent wire nets with the optimized adjacent parameters is less than or equal to the maximum crosstalk capacitance, and the difference between the maximum crosstalk capacitance and the actual crosstalk capacitance is less than or equal to a preset value.

Optionally, the method further includes:

judging whether the other nets and the adjacent nets are quasi-isomorphic;

if so, calculating to obtain an optimized adjacent parameter of the other nets according to the maximum crosstalk capacitance and the initial adjacent parameters of the other nets, so as to adjust the other nets based on the optimized adjacent parameters of the other nets, and enable the actual crosstalk capacitance existing in the other nets after adjustment to be smaller than or equal to the maximum crosstalk capacitance; the other nets include adjacent third and fourth nets.

Optionally, the determining whether the other nets and the adjacent net are quasi-isomorphic includes:

judging whether the sizes of the geometric figures of the adjacent line network and the other line networks are close to corresponding matching, whether the position relation between the geometric figures is close to corresponding matching, and whether the frequency of the borne electrical signal is close to corresponding matching;

if the judgment result is yes, determining that the adjacent net and the other nets are quasi-isomorphic;

or the like, or, alternatively,

judging whether the sizes of the geometric figures of the adjacent net and the other nets are close to corresponding matching, whether the position relation between the geometric figures is close to corresponding matching, whether the frequency of a borne electrical signal is close to corresponding matching and whether the temperature distribution is close to corresponding matching;

and if so, determining that the adjacent net and the other nets are quasi-isomorphic.

Optionally, the method further includes:

judging whether the other nets and the adjacent nets are isomorphic;

and if so, taking the optimized adjacent parameter of the adjacent net as the optimized adjacent parameter of the other net, so as to adjust the other net based on the optimized adjacent parameter of the other net, and enabling the actual crosstalk capacitance existing in the adjusted other net to be smaller than or equal to the maximum crosstalk capacitance.

Optionally, the determining whether the other nets and the adjacent net are isomorphic includes:

judging whether the geometric figures of the adjacent net and the other nets are correspondingly matched, whether the position relation between the geometric figures is correspondingly matched and whether the borne electrical signals are correspondingly matched;

if the judgment result is yes, determining that the adjacent net and the other nets have the same structure;

or the like, or, alternatively,

judging whether the geometric figures of the adjacent net and the other nets are correspondingly matched, whether the position relation between the geometric figures is correspondingly matched, whether the borne electrical signals are correspondingly matched and whether the temperature distribution is correspondingly matched;

and if so, determining that the adjacent net and the other nets have the same structure.

The embodiment of the present application further provides an adjusting apparatus for a circuit layout, including:

the adjacent net determining unit is used for determining adjacent nets in the layout to be adjusted according to the adjacent relation of the nets in the layout to be adjusted, wherein the adjacent nets comprise a first net and a second net which are adjacent;

the crosstalk judgment unit is used for simulating the circuit netlist including the adjacent nets, judging whether the adjacent nets have functional destructive crosstalk according to simulation waveforms, and activating the capacitance determination unit if the adjacent nets have the functional destructive crosstalk;

the capacitance determining unit is used for determining the maximum crosstalk capacitance without functional destructive crosstalk in the adjacent wire nets;

a first parameter calculating unit, configured to calculate an optimized neighbor parameter of the neighboring net according to the maximum crosstalk capacitance and the initial neighbor parameter of the neighboring net, so as to adjust the neighboring net based on the optimized neighbor parameter; wherein the actual crosstalk capacitance present in the neighboring nets having the optimized neighbor parameter is less than or equal to the maximum crosstalk capacitance.

Optionally, the capacitance determining unit includes:

the crosstalk capacitance adjusting unit is used for adjusting the crosstalk capacitance value of the adjacent wire mesh;

the circuit simulation unit is used for carrying out circuit simulation on the adjacent nets according to the crosstalk capacitance values;

the modeling unit is used for establishing a crosstalk capacitance-line network waveform model of the first line network and a crosstalk capacitance-line network waveform model of the second line network according to the plurality of groups of crosstalk capacitance values and corresponding circuit simulation results;

the maximum crosstalk capacitance calculating unit is used for calculating the maximum crosstalk capacitance which can be borne by the first wire network according to the crosstalk capacitance-wire network waveform model of the first wire network and calculating the maximum crosstalk capacitance which can be borne by the second wire network according to the crosstalk capacitance-wire network waveform model of the second wire network;

and the capacitance determining subunit is configured to select a minimum value of the maximum crosstalk capacitance that can be borne by the first net and the maximum crosstalk capacitance that can be borne by the second net, as a maximum crosstalk capacitance that does not have functional destructive crosstalk in the adjacent nets.

Optionally, the crosstalk determining unit includes:

the first simulation unit is used for carrying out front simulation and back simulation on the circuit netlist comprising the adjacent nets to obtain a front simulation waveform and a back simulation waveform;

the first judgment unit is used for further determining whether the front simulation waveform of the second net has waveform change corresponding to the rear simulation waveform of the first net within a preset time before the first time period if the changes of the front simulation waveform and the rear simulation waveform of the first net are inconsistent within the first time period; if so, determining whether a latch or a trigger of which the input end is connected with the first net has a change corresponding to the post-simulation waveform of the first net or not in a second time period; the second time period is a preset time period after the latest clock signal effective time point after the first time period starts; if yes, determining that the first net in the adjacent nets has functional destructive crosstalk;

or the like, or, alternatively,

the second simulation unit is used for carrying out front simulation and back simulation on the circuit netlist comprising the adjacent nets to obtain a front simulation waveform and a back simulation waveform;

the second judgment unit is used for determining a clock signal time window if the latch or the trigger of which the input end is connected with the first net generates a change inconsistent with the previous simulation waveform of the first net in a second time period; the second time period is a preset time period after the latest clock signal effective time point after the first time period starts; judging that the changes of the front simulation waveform and the rear simulation waveform of the first net in the clock signal time window are not consistent, if so, further determining whether the front simulation waveform of the second net in the clock signal time window has a waveform change corresponding to the rear simulation waveform of the first net; and if so, determining that the first net in the adjacent nets has functional destructive crosstalk.

Optionally, the initial proximity parameters include an initial parallel line length and an initial line pitch between the first net and the second net, and the first parameter calculating unit includes:

the first distance calculation unit is used for calculating to obtain a feasible connecting line distance according to the initial parallel line length and the maximum crosstalk capacitance;

a first optimized interval determining unit, configured to use the first preset interval as an optimized link interval if the feasible link interval is smaller than or equal to a first preset interval; if the feasible connecting line spacing is larger than the first preset spacing and is smaller than or equal to a second preset spacing, taking the feasible connecting line spacing as an optimized connecting line spacing;

and the first parameter determining unit is used for taking the initial parallel line length and the optimized connecting line distance as optimized adjacent parameters.

the second distance calculation unit is used for calculating to obtain a feasible connecting line distance according to the initial parallel line length and the maximum crosstalk capacitance;

a second optimized interval determining unit, configured to use the first preset interval as an optimized link interval if the feasible link interval is smaller than or equal to a first preset interval;

and the second parameter determining unit is used for calculating to obtain an optimized parallel line length according to the optimized link distance and the maximum crosstalk capacitance, and taking the optimized parallel line length and the optimized link distance as optimized adjacent parameters.

the line length calculating unit is used for calculating to obtain an optimized parallel line length according to the initial line spacing and the maximum crosstalk capacitance;

and the third parameter determining unit is used for taking the initial wiring distance and the optimized parallel line length as optimized adjacent parameters.

Optionally, the apparatus further comprises:

the capacitance calculating unit is used for calculating the initial parasitic capacitance of the adjacent net according to the initial adjacent parameters;

a second parameter calculating unit, configured to calculate an optimized neighbor parameter of the neighboring net according to the maximum crosstalk capacitance and the initial neighbor parameter of the neighboring net, if the initial parasitic capacitance is smaller than the maximum crosstalk capacitance and a difference between the maximum crosstalk capacitance and the initial parasitic capacitance is greater than a preset value, so as to adjust the neighboring net based on the optimized neighbor parameter; and the actual crosstalk capacitance existing in the adjacent wire nets with the optimized adjacent parameters is less than or equal to the maximum crosstalk capacitance, and the difference between the maximum crosstalk capacitance and the actual crosstalk capacitance is less than or equal to a preset value.

Optionally, the apparatus further comprises:

the quasi-isomorphism judging unit is used for judging whether the other nets and the adjacent nets are quasi-isomorphism or not, and if so, the third parameter calculating unit is activated;

the third parameter calculating unit is configured to calculate an optimized neighbor parameter of the other nets according to the maximum crosstalk capacitance and the initial neighbor parameters of the other nets, so as to adjust the other nets based on the optimized neighbor parameters of the other nets, and enable actual crosstalk capacitances existing in the other nets after adjustment to be smaller than or equal to the maximum crosstalk capacitance; the other nets include adjacent third and fourth nets.

Optionally, the quasi-isomorphic judging unit is specifically configured to:

or the like, or, alternatively,

Optionally, the apparatus further comprises:

the isomorphism judging unit is used for judging whether the other nets and the adjacent nets are isomorphism or not, and if so, the fourth parameter calculating unit is activated;

the fourth parameter calculating unit is configured to use the optimized neighbor parameter of the neighboring net as the optimized neighbor parameter of the other net, so as to adjust the other net based on the optimized neighbor parameter of the other net, so that the actual crosstalk capacitance existing in the other net after adjustment is smaller than or equal to the maximum crosstalk capacitance.

Optionally, the isomorphic judging unit is specifically configured to:

or the like, or, alternatively,

The embodiment of the application provides a method and a device for adjusting a circuit layout, wherein adjacent nets in the layout to be adjusted are determined according to the adjacent relation of the nets in the layout to be adjusted, the adjacent nets can comprise a first net and a second net, a circuit netlist comprising the adjacent nets is simulated, whether the adjacent nets have functional destructive crosstalk or not is judged according to a simulation waveform, if yes, the maximum crosstalk capacitance of the adjacent nets without the functional destructive crosstalk is further determined, the optimized adjacent parameters of the adjacent nets are calculated according to the maximum crosstalk capacitance and the initial adjacent parameters of the adjacent nets, the adjacent nets are adjusted based on the optimized adjacent parameters, so that the actual crosstalk capacitance existing in the adjacent nets with the optimized adjacent parameters is smaller than or equal to the maximum crosstalk capacitance, and the functional destructive crosstalk does not exist between the adjusted adjacent nets, therefore, the embodiment of the application can automatically determine the adjacent nets to be adjusted, and determining proper adjacent parameters for the adjacent nets, reducing the number of layout iterations and improving the layout design efficiency.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art according to the drawings.

FIG. 1 is a flow chart of a method for adjusting a circuit layout according to an embodiment of the present application;

fig. 2 is a block diagram of a circuit layout adjustment apparatus according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Currently, as integrated circuit fabrication processes advance, integrated circuit layouts are prone to crosstalk problems, i.e., parasitic capacitances exist between geometrically adjacent nets through which electrical signals on one net can couple to the other net, thereby disrupting the signal integrity of the other net. In severe cases, the function of the nets affected by crosstalk is damaged, which affects the function of the whole circuit, and thus, in order to ensure that the circuit system works normally, it is necessary to ensure that the crosstalk between nets does not affect the function of the nets.

Currently, it is believed that layout constraints associated with crosstalk are specified prior to designing a circuit layout, thereby driving the layout. In order to find out the functional destructive crosstalk existing in the physical layout, a circuit netlist containing parasitic parameters can be extracted from complete layout data, circuit simulation is carried out, and then circuit simulation waveforms are manually analyzed to find out a line network and a physical connection line thereof related to the functional destructive crosstalk noise. That is, it is necessary to manually identify the adjacent nets to be adjusted, and then manually determine the adjustment manner, so that it is difficult to manually and accurately determine the maximum parasitic crosstalk capacitance value that the adjacent nets can bear at one time, and it is also difficult to adjust the physical connection line at one time according to the maximum parasitic crosstalk capacitance value that can bear to meet the restrictive requirement of the parasitic capacitance, which results in a large number of iterations of the layout generated in the physical layout process, and restricts the improvement of the layout design efficiency.

Based on this, the embodiments of the present application provide a method and an apparatus for adjusting a circuit layout, wherein adjacent nets in the layout to be adjusted are determined according to a net proximity relationship in the layout to be adjusted, the adjacent nets may include a first net and a second net, a circuit netlist including the adjacent nets is simulated, whether functional destructive crosstalk exists in the adjacent nets is determined according to a simulation waveform, if yes, a maximum crosstalk capacitance without the functional destructive crosstalk in the adjacent nets is further determined, an optimized adjacent parameter of the adjacent nets is obtained by calculation according to the maximum crosstalk capacitance and an initial adjacent parameter of the adjacent nets, the adjacent nets are adjusted based on the optimized adjacent parameter, so that an actual crosstalk capacitance existing in the adjacent nets with the optimized adjacent parameter is smaller than or equal to the maximum crosstalk capacitance, and thus, the adjusted adjacent nets do not have the functional destructive crosstalk, therefore, the method and the device can automatically determine the adjacent nets to be adjusted, determine the proper adjacent parameters for the adjacent nets, reduce the number of layout iterations and improve the layout design efficiency.

The following describes a specific implementation manner of a method and an apparatus for adjusting a circuit layout according to an embodiment of the present application in detail by using the embodiments with reference to the drawings.

Referring to fig. 1, a flow chart of a method for adjusting a circuit layout according to an embodiment of the present application is shown, which may include the following steps.

S101, according to the adjacent relation of the nets in the layout to be adjusted, determining adjacent nets in the layout to be adjusted.

An integrated circuit may be an intermediate or final product in which a semiconductor material is used as a substrate and at least two or more of the active components and some or all of the interconnections are integrated in or on the substrate to perform some electronic function. Integrated circuit layouts (layout-designs) refer to the three-dimensional arrangement of two or more components and some or all of the interconnect lines of at least one active component in an integrated circuit, or the three-dimensional arrangement as prepared for the fabrication of an integrated circuit, which is, in general, a layout design that determines the geometric arrangement and connection of the electronic components used to fabricate the integrated circuit in a conductive material.

In the embodiments of the present application, the layout to be adjusted can be a three-dimensional configuration parameter of an integrated circuit prepared for manufacturing the integrated circuit, and the layout to be adjusted can include parameters of a plurality of nets and proximity parameters between the nets, such as proximity relations between physical wires of the nets, wire spacing between physical wires of adjacent nets, parallel length between physical wires of adjacent nets, and the like.

According to the layout to be adjusted, the adjacent nets in the layout to be adjusted can be determined according to the proximity relationship between the physical connection lines of the nets, and the adjacent nets can include a first net and a second net which are adjacent, and the physical connection lines of the first net and the second net have no physical connection lines of other nets in a partial area. It is understood that the number of adjacent nets in the defined layout may be one or more.

The layout to be adjusted can be a complete physical layout, a virtual physical layout obtained by a virtual prototype tool according to the circuit netlist, or a partial layout extracted from the physical layout or the virtual physical layout and at least comprising adjacent nets. When the layout to be adjusted is a partial layout, the equivalence between the adjacent nets in the partial layout and the adjacent nets in the complete layout or the virtual physical layout can be checked in advance, that is, the circuit equivalence of the nets is not considered when the parasitic effect is not considered, so as to ensure that the layout to be adjusted is an accurate layout design.

Initial proximity parameters for adjacent nets can be determined from the layout to be adjusted, and the initial proximity parameters can include initial parallel line length and initial line spacing between physical lines of the adjacent nets, i.e., initial parallel line length L between physical lines of the first net and the second net₁And an initial link spacing D₁。

Thus, based on the initial proximity parameters of the neighboring nets, the initial parasitic parameters of the neighboring nets, such as the initial parasitic capacitance C, can be calculated₁And initial parasitic resistance, etc., the initial parasitic parameters of adjacent nets may reflect the crosstalk strength between adjacent nets to some extent, such as initial parasitic capacitance C₁The larger the cross-talk between adjacent nets is. Wherein the initial parasitic capacitance C₁This can be obtained by the following formula:

where ε is the dielectric constant of the insulating medium between the physical connections of adjacent nets, T_wireIs the thickness of the metal layer of the physical connection of adjacent meshes.

S102, simulating the circuit netlist including the adjacent nets, judging whether the adjacent nets have functional destructive crosstalk according to the simulation waveforms, and if so, executing S103.

Parasitic capacitances exist between geometrically adjacent nets, and an electrical signal on one net can be coupled to another net through the parasitic capacitances, so that the integrity of the signal on the other net is damaged, for example, the signal on the other net is distorted due to the influence, and thus the parasitic capacitances in the embodiments of the present application can also be referred to as crosstalk capacitances. Strictly speaking, signal crosstalk between geometrically adjacent nets is unavoidable, but in order to ensure that a circuit system works accurately, destructive crosstalk is required to be avoided for circuit functions, that is, when parasitic capacitances of the nets are large to a certain extent, influence of an electric signal on one net on a signal of another net damages the function of the other net, and thus influences the overall function of the circuit system.

Whether the parasitic capacitance between the geometrically adjacent nets affects the function of the net can be determined according to the functional characteristics of the net, and for the net with a high requirement for the signal fineness, the smaller parasitic capacitance affects the function of the net.

Therefore, in the embodiment of the present application, a certain test stimulus can be applied to the circuit netlist including the adjacent nets for simulation, so as to determine whether the adjacent nets have the functional destructive crosstalk according to the simulation waveform. If the layout to be adjusted comprises a plurality of groups of adjacent nets, the circuit netlist comprising each group of adjacent nets can be simulated in sequence to obtain the result of whether each group of adjacent nets has functional destructive crosstalk.

The simulation of the circuit netlist including the adjacent nets to determine whether the adjacent nets have the functional destructive crosstalk can be performed in the following two ways:

in a first implementation, pre-simulation and post-simulation can be performed on the circuit netlist of the adjacent nets to obtain pre-simulation waveforms and post-simulation waveforms. If the variation of the front simulation waveform and the variation of the rear simulation waveform of the first net are inconsistent in the first time period, which indicates that the function of the first net is affected, the first time period can pass (t)₁，t₂) And (4) showing. For example, the first net pre-simulation waveform is at (t)₁，t₂) There is no waveform change in the inner, but there is a waveform change in the post-simulation waveform of the first net.

At this time, it can be further determined whether there is a waveform variation corresponding to the post-simulation waveform of the first net in the pre-simulation waveform of the second net within a preset time period before the first time period,if so, the change in the post-simulation waveform of the first net is considered to be caused by crosstalk of the second net, the first net is subjected to crosstalk, the waveform change can be defined as crosstalk noise, and the first net and the second net form a pair of a victim net and an aggressor net of the crosstalk noise. Wherein the preset time can be represented by δ, and the preset time period before the first time period can pass (t)₁-δ，t₁) And (4) showing. For example, the pre-simulation waveform for the second net is at (t)₁-δ，t₁) A waveform change occurs that corresponds to the post-simulation waveform of the first net.

The first net is connected with the input end of the latch or the trigger, and the latch or the trigger can be further determined to be at t₁Whether a waveform change different from the previous simulation waveform of the first net occurs or not within a preset time period after the effective time point of the latest clock signal after the time point, if so, the first net is considered to be latched by a latch or a trigger, so that the function of the latch or the trigger is changed, therefore, the crosstalk between the first net and the second net affects the function of the first net, and the first net is subjected to functional destructive crosstalk. Specifically, at t₁After the most recent clock signal valid time point after the time point, it can be expressed as (t)₃，t₃+ τ). Wherein, t₃Is t₁The most recent clock signal valid time point after the time point, τ is the output delay time length, that is, the preset time period. E.g. latches or flip-flops at (t)₃，t₃+ τ) a waveform change occurs corresponding to the post-simulation waveform of the first net.

In a second implementation manner, pre-simulation and post-simulation can be performed on the circuit netlists of adjacent nets to obtain pre-simulation waveforms and post-simulation waveforms.

The first net is connected with the input end of a latch or a trigger, so that whether the latch or the trigger generates a waveform change different from the pre-simulation waveform of the first net in a preset time period after the effective time point of the clock signal can be determined, and if so, the function of the first net is considered to be abnormal. A period of time after the clock signal valid time point may beTo be represented as (t)₃，t₃+ τ), where t₃And tau is the effective time point of the clock signal, and is the output delay time length, namely the preset time period.

The clock signal time window in which the waveform changes can be determined from the operating principle of the latch or flip-flop. And judging whether the front simulation waveform and the rear simulation waveform of the first net have inconsistent changes within the time window of the clock signal, if so, determining that the signal of the first net is influenced, so that the latch or the trigger is abnormal. For example, within a clock signal time window, the front simulation waveform for the first net has no waveform change, while the back simulation waveform for the first net has a waveform change.

Then, it can be determined that within the time window of the clock signal, there is a waveform change corresponding to the front simulation waveform of the second net and the rear simulation waveform of the first net, and if so, it can be considered that the waveform change of the second net has a certain influence on the waveform change of the first net, which results in a change of the function of the latch or the trigger, so that the crosstalk between the first net and the second net affects the function of the first net, and the first net is subjected to functional destructive crosstalk.

In the above two manners, it can be determined whether functional destructive crosstalk exists in the neighboring nets, and if the determination result is yes, it indicates that the crosstalk of the neighboring nets has affected the functions of the nets therein, and the neighboring parameters of the neighboring nets need to be adjusted to avoid the functional destructive crosstalk, at this time, S103 may be executed.

S103, determining the maximum crosstalk capacitance of the adjacent nets without functional destructive crosstalk.

Specifically, a plurality of adjacent parameters can be obtained by changing the initial adjacent parameters of the adjacent nets, and whether the adjacent nets are functionally damaged due to crosstalk under the plurality of adjacent parameters is determined corresponding to a plurality of parasitic capacitances of the adjacent nets, so that the maximum parasitic capacitance of the adjacent nets without functional destructive crosstalk is found, that is, the maximum crosstalk capacitance C that the adjacent nets can tolerate is found₂. Once the actual crosstalk capacitance in the adjacent net is greater than the maximum crosstalk capacitance C₂The function of at least one of the adjacent nets will be destroyed.

Specifically, the maximum crosstalk capacitance in the neighboring nets where there is no functional destructive crosstalk can be determined by:

adjusting the adjacent parameters of the adjacent nets to adjust the crosstalk capacitance values of the adjacent nets, and performing circuit simulation on the adjacent nets under the multiple groups of crosstalk capacitances to obtain a circuit simulation result, wherein the circuit simulation result is a net waveform; according to the multiple groups of crosstalk capacitance values and corresponding circuit simulation results, a crosstalk capacitance-line network waveform model can be established in a curve fitting or machine learning mode, and specifically, a crosstalk capacitance-line network waveform model of a first line network and a crosstalk capacitance-line network waveform model of a second line network can be established; according to the crosstalk capacitance-net waveform model of the first net, the maximum crosstalk capacitance C which can be borne by the first net can be calculated_2，1-maxAccording to the crosstalk capacitance-net waveform model of the second net, the maximum crosstalk capacitance C which can be borne by the second net can be calculated_2，2-max(ii) a Selecting the minimum value of the maximum crosstalk capacitance which can be borne by the first net and the maximum crosstalk capacitance which can be borne by the second net as the maximum crosstalk capacitance which does not have functional destructive crosstalk in the adjacent nets, namely C₂＝min(C_2，1-max，C2，_2-max)。

And S104, calculating to obtain the optimized adjacent parameters of the adjacent nets according to the maximum crosstalk capacitance and the initial adjacent parameters of the adjacent nets.

As a possible implementation, the initial parallel line length L may be based on₁And a maximum crosstalk capacitance C₂Calculating to obtain the optimized link interval D₂At this time, the link pitch D is optimized₂And an initial parallel line length L₁The proximity parameters are configured such that the parasitic capacitance of adjacent nets is less than or equal to the maximum crosstalk capacitance C₂Therefore, the first net and the second net both function normally, so that the initial parallel line length L can be increased₁And optimizing the link pitch D₂As an optimized proximity parameter.

Specifically, the optimized link pitch D can be calculated according to the following formula₂Minimum value of D_2minTo determine an optimized inter-wire distance D₂：

As an example, according to the initial parallel line length L₁And a maximum crosstalk capacitance C₂Calculated optimized link distance D₂Comprises the following steps:

as another possible implementation, the initial connection interval D can be used₁And a maximum crosstalk capacitance C₂And calculating to obtain the optimized parallel line length L₂At this time, the initial inter-wire distance D₁And optimizing the parallel line length L₂The proximity parameters are configured such that the parasitic capacitance of adjacent nets is less than or equal to the maximum crosstalk capacitance C₂So that both the first net and the second net function properly, and so that the initial wire spacing D will be₁And optimizing the parallel line length L₂As an optimized proximity parameter.

Specifically, the optimized parallel line length L can be calculated according to the following formula₂Minimum value L of_2minTo determine the optimized parallel line length L₂：

Or the like, or, alternatively,

where ε is the dielectric constant of the insulating medium between the physical connections of adjacent nets, T_wireIs a neighboring netThickness of the metal layer of the physical connection, C₁Is the initial parasitic capacitance of the adjacent net.

As an example, based on the initial inter-wire distance D₁And a maximum crosstalk capacitance C₂Calculated optimized parallel line length L₂Comprises the following steps:

or the like, or, alternatively,

as still another possible implementation, the initial parallel line length L can be determined₁And a maximum crosstalk capacitance C₂Calculating to obtain a feasible link distance D₃At this time, the feasible connection interval D₃And an initial parallel line length L₁The proximity parameters are configured such that the parasitic capacitance of adjacent nets is less than or equal to the maximum crosstalk capacitance C₂Thus, both the first net and the second net function properly.

Considering the actual manufacturing process of the physical connection line, the first predetermined distance D may be set in the design rule₄Is the minimum spacing value between the physical connecting lines.

If feasible, connecting line spacing D₃Is smaller than the first preset distance D₄Illustrate the feasible wire spacing D₃If the distance D is too small, the short circuit between the physical connecting lines is easy to occur, and at the moment, the first preset distance D can be adjusted₄As an optimized link pitch D₂Will be the initial parallel line length L₁And optimizing the link pitch D₂As an optimized proximity parameter, i.e. the initial parallel line length L₁And a first preset distance D₄As an optimized proximity parameter.

Or according to the optimized link spacing D₂And a maximum crosstalk capacitance C₂And calculating to obtain the optimized parallel line length L₂Will optimize the wire spacing D₂And optimizing the parallel line length L₂As an optimized proximity parameter. In fact, the optimization of the parallel line length is based on the maximum crosstalk capacitance C₂And a first preset roomDistance D₄Obtained differently from the initial parallel line length L₁。

If feasible, connecting line spacing D₃Is greater than or equal to the first preset distance D₄Illustrate the feasible wire spacing D₃Satisfy the design rule, the feasible line spacing D can be obtained₃As an optimized link pitch D₂Will be the initial parallel line length L₁And optimizing the link pitch D₂As an optimized proximity parameter, i.e. the initial parallel line length L₁And a feasible link pitch D₃As an optimized proximity parameter.

In layout design, a second predetermined pitch may be set in order to limit the wire spacing between adjacent nets. If the feasible wire spacing D between adjacent nets is small₃Less than or equal to the second predetermined spacing may be considered to satisfy the design rule; if the feasible wire spacing D between adjacent nets is small₃If the distance between the two connecting lines is larger than the second preset distance, the distance between the two connecting lines is considered to be too large, so that the area of the layout is wasted, the two connecting lines are not adjacent to each other, namely, other physical connecting lines can be inserted between the two physical connecting lines, and the waste of the area of the layout is reduced as much as possible on the premise of avoiding the functional destructive crosstalk. The second preset distance may be determined according to actual conditions, and as an example, the second preset distance may be (2 × D)₄+D₅) Wherein D is₅Is the minimum width value of the physical connection line.

In an embodiment of the application, the layout to be adjusted may be adjusted based on the optimized proximity parameter. Because the optimized adjacent parameters are obtained by optimizing the initial adjacent parameters according to the maximum crosstalk capacitance, and the actual crosstalk capacitance existing in the adjacent nets with the optimized adjacent parameters is less than or equal to the maximum crosstalk capacitance, no functional damage is generated in the adjacent nets with the optimized adjacent parameters due to crosstalk, so that no functional damage is caused by crosstalk between the adjacent nets in the adjusted layout to be adjusted.

In the embodiments of the present application, there may be a problem that crosstalk is avoided and the neighboring parameters are too strict, for example, there is too large a wire spacing between the physical wires of two nets or too small a parallel length, which may result in a waste of layout area.

Based on this, if the initial parasitic capacitance of the adjacent wires is much smaller than the maximum crosstalk capacitance, it indicates that the adjacent wires have stricter proximity parameters, resulting in waste of layout area.

Therefore, when the initial parasitic capacitance is judged to be smaller than the maximum crosstalk capacitance and the difference between the maximum crosstalk capacitance and the initial parasitic capacitance is larger than the preset value, the optimized adjacent parameter can be obtained by calculation according to the initial adjacent parameter and the maximum crosstalk capacitance, so that whether the difference between the maximum crosstalk capacitance and the actual capacitance existing in the adjacent line network with the optimized adjacent parameter is smaller than the preset value or not can be determined. The manner in which the optimized proximity parameter is calculated from the initial proximity parameter and the maximum crosstalk capacitance may refer to the above implementation. In this way, the optimized proximity parameters in the embodiments of the present application do not cause too much area waste in the layout to be adjusted, compared to manually specifying the proximity parameters.

In the layout to be adjusted, there may be not only one neighboring net, but after the adjustment strategy of one neighboring net is determined, the adjustment strategies of other neighboring nets can also be determined according to the relationship between the neighboring net and other neighboring nets, so that excessive repeated data processing is not necessary.

Specifically, if the other nets include a third net and a fourth net, it may be determined whether the other nets and the adjacent net are isomorphic, and if so, the adjustment policy for the adjacent net may be reused in the other nets. For example, the optimized neighbor parameter of the neighboring net can be used as the optimized neighbor parameter of the other net, and then the other net is adjusted based on the optimized neighbor parameter of the other net, so that the actual crosstalk capacitance existing in the adjusted other net is smaller than or equal to the maximum crosstalk capacitance. Therefore, simulation and calculation of the third network and the fourth network are not needed, the data processing amount is reduced, and the layout adjustment time is saved.

Judging whether the adjacent wire nets and other wire nets are isomorphic, specifically judging whether the sizes of the geometric figures of the adjacent wire nets and other wire nets are correspondingly matched, judging whether the position relations between the geometric figures are correspondingly matched, and judging whether the frequencies of the electrical signals borne by the two geometric figures are correspondingly matched. And if the judgment results are yes, determining that the adjacent net and other nets have the same structure.

Judging whether the adjacent wire mesh and other wire meshes are isomorphic, and specifically judging whether the sizes of the geometric figures of the adjacent wire mesh and other wire meshes are correspondingly matched, judging whether the position relationship between the geometric figures is correspondingly matched, judging whether the frequencies of the electrical signals borne by the two wires are correspondingly matched, and judging whether the temperature distributions of the two wires are correspondingly matched. And if the judgment results are yes, determining that the adjacent net and other nets have the same structure.

For example, it can be determined whether the geometric figures of the adjacent net and the other nets have the same size, whether the position relationship between the geometric figures is the same, whether the frequencies of the electrical signals carried by the two nets are the same, and whether the temperature distributions of the two nets are the same, if so, it is determined that the adjacent net and the other nets have the same structure.

Specifically, if the other nets include the third net and the fourth net, it may be determined whether the other nets and the adjacent net are quasi-isomorphic, that is, whether the other nets and the adjacent net have similar structures and similar parameters. For example, the maximum crosstalk capacitance of the adjacent net can be used as the maximum crosstalk capacitance of the other net, and then the optimized proximity parameters of the other net are calculated according to the maximum crosstalk capacitance of the other net and the initial proximity parameters of the other net, and the other net is adjusted based on the optimized proximity parameters of the other net, so that the actual crosstalk capacitance of the adjusted other net is smaller than or equal to the maximum crosstalk capacitance. Therefore, simulation of other nets is not needed, data processing amount is reduced, and layout adjustment time is saved.

Whether the adjacent net and other nets are quasi-isomorphic is judged, and specifically, whether the sizes of the geometric figures of the adjacent net and other nets are close to corresponding matching or not is judged, whether the position relation between the geometric figures is close to corresponding matching or not is judged, and whether the frequencies of the electrical signals borne by the two nets are close to corresponding matching or not is judged. And if the judgment results are yes, determining that the adjacent net and other nets have the same structure.

Whether the adjacent wire nets and other wire nets are isomorphic is judged, and specifically, whether the sizes of the geometric figures of the adjacent wire nets and other wire nets are close to corresponding matching or not is judged, whether the position relation between the geometric figures is close to corresponding matching or not is judged, whether the frequencies of the electrical signals borne by the two wires are close to corresponding matching or not is judged, and whether the temperature distribution of the two wires is close to corresponding matching or not is judged. And if the judgment results are yes, determining that the adjacent net and other nets have the same structure.

For example, it can be determined whether the geometric figures of the adjacent net and the other nets are similar in size, whether the position relationship between the geometric figures is similar, whether the frequencies of the electrical signals carried by the two nets are similar, and whether the temperature distributions of the two nets are similar, if so, it is determined that the adjacent net and the other nets are isomorphic.

The judging whether the sizes of the geometric figures are similar can be specifically that whether the difference value of the sizes of the corresponding positions of the geometric figures is smaller than or equal to a preset size value is judged; judging whether the position relations of the geometric figures are similar can be specifically that whether the similarity of the corresponding positions of the position relations of the geometric figures is smaller than or equal to a position preset value is judged; judging whether the frequencies of the electrical signals carried by the two are similar can be specifically that whether the difference value of the frequencies of the electrical signals carried by the two is smaller than or equal to a preset frequency value is judged; the determining whether the temperature distributions of the two are similar may specifically be determining whether the similarity of the temperature distributions of the two is less than or equal to a preset temperature value.

The embodiment of the application provides an adjusting method of a circuit layout, which comprises the steps of determining adjacent nets in the layout to be adjusted according to the adjacent relation of the nets in the layout to be adjusted, wherein the adjacent nets can comprise a first net and a second net, simulating a circuit netlist comprising the adjacent nets, judging whether the adjacent nets have functional destructive crosstalk according to a simulation waveform, if so, further determining the maximum crosstalk capacitance which does not have the functional destructive crosstalk in the adjacent nets, calculating the optimized adjacent parameters of the adjacent nets according to the maximum crosstalk capacitance and the initial adjacent parameters of the adjacent nets, adjusting the adjacent nets based on the optimized adjacent parameters to ensure that the actual crosstalk capacitance which exists in the adjacent nets with the optimized adjacent parameters is smaller than or equal to the maximum crosstalk capacitance, so that the functional destructive crosstalk does not exist between the adjusted adjacent nets, and the embodiment of the application can automatically determine the adjacent nets to be adjusted, and determining proper adjacent parameters for the adjacent nets, reducing the number of layout iterations and improving the layout design efficiency.

Based on the above method for adjusting a circuit layout, an embodiment of the present application further provides an apparatus for adjusting a circuit layout, and referring to fig. 2, the apparatus for adjusting a circuit layout provided in an embodiment of the present application includes:

Optionally, the capacitance determining unit includes:

Optionally, the crosstalk determining unit includes:

or the like, or, alternatively,

Optionally, the apparatus further comprises:

Optionally, the quasi-isomorphic judging unit is specifically configured to:

or the like, or, alternatively,

Optionally, the apparatus further comprises:

Optionally, the isomorphic judging unit is specifically configured to:

or the like, or, alternatively,

The embodiment of the application provides an adjusting device of a circuit layout, which determines adjacent nets in the layout to be adjusted according to the adjacent relationship of the nets in the layout to be adjusted, wherein the adjacent nets can comprise a first net and a second net, the circuit netlist comprising the adjacent nets is simulated, whether the adjacent nets have functional destructive crosstalk is judged according to a simulation waveform, if yes, the maximum crosstalk capacitance of the adjacent nets without the functional destructive crosstalk is further determined, the optimized adjacent parameters of the adjacent nets are obtained by calculation according to the maximum crosstalk capacitance and the initial adjacent parameters of the adjacent nets, the adjacent nets are adjusted based on the optimized adjacent parameters, so that the actual crosstalk capacitance existing in the adjacent nets with the optimized adjacent parameters is smaller than or equal to the maximum crosstalk capacitance, and the functional destructive crosstalk does not exist between the adjusted adjacent nets, therefore, the embodiment of the application can automatically determine the adjacent nets to be adjusted, and determining proper adjacent parameters for the adjacent nets, reducing the number of layout iterations and improving the layout design efficiency.

The name "first" in the names "first … …", "first … …", etc. mentioned in the embodiments of the present application is only used for name identification, and does not represent the first in sequence. The same applies to "second" etc.

As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the above embodiment methods can be implemented by software plus a general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a storage medium, such as a read-only memory (ROM)/RAM, a magnetic disk, an optical disk, or the like, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the method according to the embodiments or some parts of the embodiments of the present application.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. The above-described embodiments of the apparatus and system are merely illustrative, wherein modules described as separate parts may or may not be physically separate, and parts shown as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only a preferred embodiment of the present application and is not intended to limit the scope of the present application. It should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the scope of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims

1. A method of adjusting a circuit layout, comprising:

2. The method of claim 1, wherein said determining a maximum crosstalk capacitance for which functional destructive crosstalk is not present in said neighboring nets comprises:

3. The method of claim 1, wherein said simulating the circuit netlist including the neighboring net and determining whether the neighboring net has functional destructive crosstalk according to the simulation waveforms of the neighboring net comprises:

or the like, or, alternatively,

4. The method of claim 1, wherein the initial proximity parameters comprise an initial parallel wire length and an initial wire spacing between the first net and the second net, and wherein calculating an optimized proximity parameter based on the maximum crosstalk capacitance and the initial proximity parameters comprises:

5. The method of claim 1, wherein the initial proximity parameters comprise an initial parallel wire length and an initial wire spacing between the first net and the second net, and wherein calculating an optimized proximity parameter based on the maximum crosstalk capacitance and the initial proximity parameters comprises:

6. The method of claim 1, wherein the initial proximity parameters comprise an initial parallel wire length and an initial wire spacing between the first net and the second net, and wherein calculating an optimized proximity parameter based on the maximum crosstalk capacitance and the initial proximity parameters comprises:

7. The method of any one of claims 1-6, further comprising:

8. The method of any one of claims 1-6, further comprising:

judging whether the other nets and the adjacent nets are quasi-isomorphic;

9. The method of claim 8, wherein said determining whether said other net is quasi-homogeneous with said neighboring net comprises:

or the like, or, alternatively,

10. The method of any one of claims 1-6, further comprising:

judging whether the other nets and the adjacent nets are isomorphic;

11. The method of claim 10, wherein said determining whether said other net is homogeneous with said neighboring net comprises:

or the like, or, alternatively,

12. An apparatus for adjusting a circuit layout, comprising: