JP2007018281A - Processing method of cad data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a processing method of CAD data by which a user of a CAD system can easily check whether the CAD data are correct or not. <P>SOLUTION: CAD data in which a plurality of pattern data are superimposed, and process management information including a plurality of process names are read in a CAD system (step A1). Then pattern data corresponding to individual step names are checked (step A2). Also, at least the maximum width yi of the second coordinate axis of each pattern data is computed (step A3). Next, one process name is selected and the pattern data corresponding to it is shifted equivalent to the shifting width obtained by adding predetermined distance α to the maximum width yi and is displayed on a display screen 10 along with the process name (steps A4-A6). When the pattern data corresponding to selected process name does not exist, the process name is indicated by flashing (step A7). These steps A4-A7 are repeated until selected all process names are completed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a semiconductor device automatic design method using a CAD system, and more particularly to a CAD data processing method for displaying CAD data in which a plurality of pattern data overlap on a screen.

  In recent years, along with the increase in the scale of semiconductor integrated circuits, the tendency of high density and multi-layered cells and wirings constituting the semiconductor integrated circuit is increasing. Such a semiconductor integrated circuit is designed through automatic design by a CAD (Computer Aided Design) system using an electronic computer, that is, functional design, logical design, layout design, and the like. In the function design, a desired semiconductor integrated circuit function is specified by a hardware description language, and in the logic design, logic circuit data is generated by performing logic synthesis so as to realize the function specified in the function design. In layout design, a logic circuit is laid out on a semiconductor chip based on logic circuit data logically synthesized in the logic design.

  In the layout design described above, not only the circuit layout but also various marks formed on the semiconductor wafer in the actual manufacturing process, such as process management marks for recognizing a plurality of different processes, alignment marks, etc. Is done. These marks are described on the CAD system by a plurality of pattern data (hereinafter referred to as “pattern data”) corresponding to a coordinate system composed of first and second coordinate axes (that is, a two-dimensional Cartesian coordinate system). Do). The plurality of pattern data are superimposed on each other in the coordinate system and stored in a storage area such as a database as one CAD data.

In addition, as a related technical document, the following patent documents are mentioned, for example.
Japanese Patent Laid-Open No. 2004-288685

  However, in the above-described automatic design of the semiconductor integrated circuit according to the conventional example, as shown in FIG. 5, when the CAD data 30 is displayed on the display screen 10, a plurality of pattern data 40 are displayed in an overlapping state. It had been. At this time, it is possible to extract one pattern data 40 from a plurality of pattern data 40 that overlap each other, but only one extracted pattern data 40 is displayed on the display screen 10. For this reason, it is extremely difficult to confirm the correctness of the arranged pattern data 40 and to confirm the arrangement relationship between different pattern data 40 in the coordinate system. As a result, the burden on the user of the CAD system has increased in the automatic design of the semiconductor integrated circuit.

  Therefore, the present invention provides a CAD data processing method that allows a user of a CAD system to easily check whether CAD data is correct or incorrect.

  The CAD data processing method of the present invention has been made in view of the above-described problems. In a CAD system for automatically designing a semiconductor device, a plurality of pattern data corresponding to the first and second coordinate axes overlap. The CAD data processing method for displaying the CAD data on the screen, the step of reading the CAD data and the process management information consisting of a set of a plurality of process names into the CAD system, and from the CAD data to each process name Extracting corresponding pattern data and calculating the maximum width on the second coordinate axis, and each pattern data not overlapping each other along the second coordinate axis with each shift width based on each maximum width And sequentially displaying each process name corresponding to each pattern data on the screen.

  According to the CAD data processing method of the present invention, it is possible to easily confirm CAD data in which a plurality of pattern data are superimposed in automatic design of a semiconductor integrated circuit by a CAD system. That is, it is possible to significantly reduce the burden on the user of the CAD system when automatically designing a semiconductor integrated circuit.

  Next, a CAD data processing method according to an embodiment of the present invention will be described. Here, the CAD data means CAD data used for layout design among automatic designs of a semiconductor integrated circuit performed by a CAD system in which a predetermined program is installed in an electronic computer. The CAD data is formed by superimposing a plurality of pattern data corresponding to a coordinate system consisting of first and second coordinate axes (that is, a two-dimensional Cartesian coordinate system). It is assumed that the mark is a process management mark for recognizing a plurality of different processes.

  The pattern data may be marks other than those described above as long as they are various marks formed on a semiconductor wafer or a layer stacked thereon in an actual process. For example, the pattern data may be alignment marks or the like.

  Next, a flow of a CAD data processing method according to the present embodiment and a display screen on which CAD data is displayed will be described with reference to the drawings. 1 and 2 are flowcharts showing a CAD data processing method according to this embodiment. FIG. 1 shows the main routine, and FIG. 2 shows the subroutine of FIG. FIG. 3 is a diagram showing pattern data extracted from CAD data according to the present embodiment. FIG. 4 shows a display screen of CAD data according to the present embodiment. 3 and 4, the same components as those shown in FIG. 5 are described with the same reference numerals.

  As shown in the main routine of FIG. 1, in step A1, CAD data is read into the CAD system. In addition, process management information including process names of a plurality of processes necessary for the manufacturing process is read into the CAD system. Here, the plurality of process names are assumed to be N process names. The reading of the CAD data or the process management information is performed by a user operation using a so-called redirection function (data file input / output function by the operating system) or the like, but is automatically performed by a program. May be.

  Next, in step A2, the pattern data corresponding to each process name is searched with reference to N process names included in the process management information and a plurality of pattern data constituting CAD data. A set of process names and corresponding pattern data is stored in the first storage area. If there is no pattern data corresponding to the process name, only the process name is stored in the first storage area.

  Next, in step A3, each maximum width xi (i = 1, 2,... N) of each pattern data on the first coordinate axis and each maximum width yi (i = 1, 2,... N) on the second coordinate axis. ... N) is calculated. Here, each maximum width xi, yi of the pattern data is, in other words, each of the minimum rectangles (so-called boundary boxes) including all coordinate points of the pattern data along the first and second coordinate axes. The length of the side.

  The calculation of the maximum widths xi and yi on the first and second coordinate axes is performed, for example, in a subroutine of FIG. That is, in step B1, one process name is selected from N process names, and pattern data corresponding to it is extracted. In step B2, the maximum width xi in the first direction of the pattern data is calculated. The calculation is performed as follows, for example.

  First, for each coordinate point included in the pattern data, the distance from the coordinate origin on the first coordinate axis is calculated. Next, as shown in FIG. 3A, the minimum value xia and the maximum value xib are searched from these distances, and the absolute value of the difference between them is calculated. This absolute value is taken as the maximum width xi on the first coordinate axis.

  Next, in step B3, the maximum width yi in the second direction of the pattern data corresponding to the process name selected in step B1 is calculated. The calculation is performed as follows, for example, similarly to the maximum width xi in the first direction.

  First, for each coordinate point included in the pattern data, the distance from the coordinate origin on the second coordinate axis is calculated. Next, as shown in FIG. 3B, the minimum value yia and the maximum value yib are searched from these distances, and the absolute value of the difference between them is calculated. This absolute value is taken as the maximum width yi on the second coordinate axis.

  The calculation of the maximum widths xi, yi of the pattern data described above is repeated until all process names are selected, and the maximum widths xi, yi of all the pattern data are calculated. When there is no pattern data corresponding to the selected process name, for example, the maximum widths xi and yi are set to zero as so-called NULL data. The maximum widths xi and yi on the first and second coordinate axes of each pattern data calculated in this way are stored in a second storage area such as a buffer.

  Next, in step A4 of the main routine of FIG. 1, one process name is selected from a plurality of process names included in the process management information. The selection of the process name is preferably performed according to the process sequence of the manufacturing process. Then, with reference to the first storage area, pattern data corresponding to the selected process name is selected, and only the pattern data is extracted from the CAD data.

  Then, as shown in FIG. 4, when there is pattern data corresponding to the selected process name (that is, when the maximum widths xi and yi are not zero), the pattern data 40 corresponding to the process name is displayed on the display screen. 10 pattern data display areas 11 are displayed. When displaying the process name and the pattern data 40, in step A5 and step A6, referring to the second storage area, a predetermined distance α is added to the maximum width yi of the second coordinate axis of the pattern data 40. The pattern data 40 is shifted and displayed along the second coordinate axis by the shift width Ls. Here, the predetermined distance α is an arbitrary distance so that even if a plurality of process names and pattern data 40 are repeatedly displayed along the second coordinate axis, a shift width Ls is obtained such that they do not overlap each other. Set by. Further, the pattern data 40 is displayed while maintaining the original coordinates without being shifted on the first coordinate axis.

  Although not shown, when the pattern data 40 is displayed, the pattern data 40 is shifted and displayed along the first coordinate axis with reference to the maximum width xi on the first coordinate axis, as necessary. May be.

  On the other hand, if there is no pattern data corresponding to the selected process name, in step A7, the maximum value ymax is searched from the maximum widths yi on the second coordinate axis with reference to the second storage area, The maximum value ymax is defined as a shift width Ls. Alternatively, the predetermined distance α is added to the maximum value ymax, and this is set as the shift width Ls. Then, the process name is shifted and displayed along the second coordinate axis with the shift width Ls. At this time, the process name is displayed in a blinking manner, for example, so as to clearly indicate to the user that the corresponding pattern data does not exist. Further, in the area where the pattern data 40 is originally displayed, the range of the shift width Ls is a non-display area where nothing is displayed. Thereby, the visibility of the process name which does not have pattern data can be improved reliably.

  Steps A4 to A7 are repeated N times until all process names are selected. In this way, as shown in FIG. 4, each process and each pattern data 40 corresponding thereto can be displayed in parallel at the same time without overlapping each other.

  The CAD data in which the plurality of pattern data 40 overlap may be displayed in a predetermined area of the display screen 10, for example, the CAD data display area 12. Alternatively, in the CAD data display area 12, when the read CAD data includes pattern data that does not correspond to any process name of the read process management information, only the pattern data may be displayed. . In this case, every time the pattern data 40 corresponding to the process name selected in step A4 is extracted from the CAD data, the pattern data 40 is deleted from the CAD data, and the remaining pattern data is displayed in the CAD data display area 12. You can do it.

  Thus, according to the present embodiment, each process name and each pattern data 40 corresponding thereto can be displayed in parallel so as not to overlap each other along the second coordinate axis. Therefore, when the user of the CAD system confirms the correctness of the plurality of pattern data 40 constituting the CAD data, extraction and display of the pattern data 40 corresponding to one process name is performed as in the conventional example. It is not necessary to repeat the process name. That is, it is easy for the CAD system user to check whether the CAD data is correct.

  Moreover, when the 1st coordinate axis of the pattern data 40 displayed on the display screen 10 is being fixed, the user can confirm easily the arrangement | positioning relationship between each pattern data 40 regarding a 1st coordinate axis. As a result, it is possible to greatly reduce the burden on the user of the CAD system in the automatic design of the semiconductor integrated circuit.

  In the above-described embodiment, the step of calculating the first and second maximum widths of the pattern data (steps B2 and B3 in FIG. 2) is performed in the subroutine of FIG. 2 as step A3. Is not limited to this. That is, step B2 and step B3 for calculating the first and second maximum widths of the pattern data may be performed after step A4 in the loop of the main routine of FIG.

  In the above embodiment, the maximum widths xi and yi on the first and second coordinate axes are calculated. However, the present invention is not limited to this. That is, if it is not necessary to shift and display the pattern data along the first coordinate axis, the step B2 for calculating the maximum width xi on the first coordinate axis may be omitted.

It is a flowchart explaining the processing method of CAD data which concerns on embodiment of this invention. It is a flowchart explaining the processing method of CAD data which concerns on embodiment of this invention. It is a figure which shows the pattern data extracted from CAD data based on embodiment of this invention. It is a figure which shows the display screen of CAD data which concerns on embodiment of this invention. It is a figure which shows the display screen of CAD data which concerns on a prior art example.

Explanation of symbols

10 Display screen 11 Pattern data display area 12 CAD data display area 30 CAD data 40 Pattern data

Claims (5)

In a CAD system for automatically designing a semiconductor device, a CAD data processing method for displaying on a screen CAD data in which a plurality of pattern data corresponding to first and second coordinate axes are overlapped,
Reading the CAD data and process management information comprising a plurality of process names into the CAD system;
Extracting each pattern data corresponding to each process name from the CAD data, and calculating each maximum width of each pattern data in the second coordinate axis;
Each pattern data is shifted on the screen along with the process names corresponding to each pattern data while sequentially shifting the pattern data so as not to overlap each other with the respective shift widths based on the respective maximum widths. And a step of displaying the CAD data. Calculating the maximum width of the pattern data on the second coordinate axis;
For each coordinate point included in the pattern data, calculating a distance from the coordinate origin in the second coordinate axis;
The CAD data processing method according to claim 1, further comprising: calculating a difference between the minimum value and the maximum value of the distance and determining an absolute value of the difference as the maximum width. 3. The CAD data according to claim 1, further comprising a step of blinking and displaying the process name on the screen when pattern data corresponding to the process name does not exist in the CAD data. Processing method. 4. The CAD data processing method according to claim 1, wherein the pattern data is a mark for process management. 5. The step of displaying the pattern data, wherein each pattern data is displayed by being shifted not only along the second coordinate axis but also along the first coordinate axis. A CAD data processing method according to any one of the above.

Images