JP6260179B2 - Arithmetic apparatus and arithmetic method - Google Patents

Description

  The present invention relates to a symbolic execution technique.

  When there are conditional branches in the program, there are a plurality of processing paths in the program. When testing a program, the tester designs test data and scenarios so that a plurality of processing paths in the program can be executed comprehensively.

  One method of testing a program is known as symbolic execution. Symbolic execution is a technique in which an initial value of a variable is set to an indeterminate symbol value instead of a real value, and a program is executed in a pseudo manner so as to cover a plurality of processing paths in the program. By performing symbolic execution, it is possible to obtain a path condition representing a history of following conditional branches for each of a plurality of processing paths in the program. Therefore, the test data of the program can be obtained by obtaining the value of the variable that satisfies the path condition for each path.

JP-A-11-272503 JP-A-6-290039 Japanese Unexamined Patent Publication No. 2011-191985 Japanese Patent Publication No. 61-5169

Corina SPas.areanu, Peter C. Mehlitz, David H. Bushnell, Karen Gundy-Burlet, Michael Lowry, Suzette Person, Mark Pape, "Combining unit-level symbolic execution and system-level concrete execution for testing nasa software", the 2008 international symposium on Software testing and analysis, Pages 15-26.

  By the way, there are various formats (also referred to as internal formats) for variables to be subjected to operations in the program (for example, transcription operation or comparison operation). For example, a programming language called COBOL has a format in which 1 digit is assigned to 1 byte (hereinafter referred to as an external decimal format), a format in which 2 digits are assigned to 1 byte (hereinafter referred to as an internal decimal format), and a value of 2 There are formats for storing in hexadecimal expression (hereinafter referred to as internal binary format).

  However, in the prior art, sufficient consideration has not been given to the handling of variable formats in symbolic execution. Therefore, with the conventional symbolic execution, a correct calculation result cannot be obtained, and as a result, test data may not be extracted without omission.

  Accordingly, an object of the present invention, in one aspect, is to provide a technique for appropriately performing variable operations in symbolic execution.

  The arithmetic device according to the present invention converts the specific value of the first variable that is the object of the symbolic execution into the first format that has a value for each digit, and the symbol value of the first variable. Is converted into a second format which is a format having a value for each digit, the specific value after conversion by the first conversion unit is stored in the first storage area, and the first conversion unit And a first storage processing unit for storing the symbol value after conversion in the second storage area different from the first storage area.

  In one aspect, variable operations can be appropriately performed in symbolic execution.

FIG. 1 is a diagram illustrating an example of a program that is a symbolic target. FIG. 2 is a diagram illustrating an example of data stored in the main storage device when the method according to the present embodiment is not used. FIG. 3 is a functional block diagram of the information processing apparatus. FIG. 4 is a functional block diagram of the symbol value calculation device. FIG. 5 is a diagram illustrating an example of data stored in the main storage device. FIG. 6 is a diagram illustrating an example of data stored in the external register. FIG. 7 is a diagram showing a main processing flow. FIG. 8 is a diagram illustrating a processing flow of a specific value transfer operation. FIG. 9 is a diagram illustrating a processing flow of a symbol value transfer operation. FIG. 10 is a diagram showing a processing flow of processing for conversion to the digit number format. FIG. 11 is a diagram illustrating an example of specific values stored in the data register by the transfer operation. FIG. 12 is a diagram illustrating an example of symbol values stored in the data register by the transfer operation. FIG. 13 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 14 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 15 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 16 is a diagram illustrating an example of specific values stored in the data register by the transfer operation. FIG. 17 is a diagram illustrating an example of symbol values stored in the data register by the transfer operation. FIG. 18 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 19 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 20 is a diagram illustrating an example of specific values stored in the data register by the transfer operation. FIG. 21 is a diagram illustrating an example of symbol values stored in the data register by the transfer operation. FIG. 22 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 23 is a diagram illustrating an example of data stored in the main storage device by the transfer operation. FIG. 24 is a diagram illustrating a processing flow of the comparison operation. FIG. 25 is a diagram illustrating a process flow of the mixed value generation process. FIG. 26 is a diagram illustrating an example of data stored in the external register. FIG. 27 is a diagram illustrating a specific example of generation of a mixed value. FIG. 28 is a diagram illustrating a specific example of generation of a mixed value.

  FIG. 1 shows an example of a program to be symbolized. The program shown in FIG. 1 is a program written in COBOL and has the following characteristics.

(1) There are a plurality of variables having different internal formats. The variable A secures an area of the main storage device in the external decimal format. The variable B reserves an area of the main storage device in the internal decimal format. The variable C secures an area of the main storage device in an internal binary format.

(2) A situation occurs in which symbol values and specific values are mixed for the same variable. Specifically, in the 110th line of the program, only the second digit of the variable A is rewritten to the specific value “4”. In the 120th line, only the third digit of the variable A is rewritten to the specific value “9”. On the other hand, the other digits are not concrete values but symbol values and the like.

(3) A value including both a symbol value and a concrete value stored in a variable is transferred to another variable by a transfer operation, and compared with another variable by a comparison operation. Specifically, in line 140, the value stored in variable A is transferred to variable B, and in line 150, the value stored in variable A is transferred to variable C. On lines 160 and 190, the value stored in variable A is compared with the values stored in variable B and variable C.

  When performing symbolic execution, “IF (variable A = variable B) THEN” on the 160th line should be true or false to cover the processing path. However, when the method of the present embodiment described below is not used, the “MOVE variable A TO variable B” on the 140th line and the “MOVE variable A TO variable C” on the 150th line are transferred, for example, as shown in FIG. The area of the main storage device is rewritten. Specifically, for variable A, specific values are stored at addresses A2 and A3, and symbol values and the like are stored at other addresses. For variable B, a specific value is stored at address A5, and a symbol value or the like is stored at other addresses. For variable C, a specific value is stored at address A8, and a symbol value or the like is stored at other addresses. Note that the position of the decimal point is separately stored in a register.

  Accordingly, in “IF (variable A = variable B) THEN” on the 160th line, it is determined whether “[00] [49] [0_0] = [49] [00]” is satisfied, so only true is satisfied. Result. In this example, the parentheses represent 1 byte. For example, if [00], the value “00” is stored in 1 byte. The last value is the value of the sign part, and in the present embodiment, there are three values: a positive value, a negative value, and no value.

  Similarly, “IF (variable A = variable C) THEN” on the 190th line should be true or false in symbolic execution. However, when this embodiment is not used, since it is determined whether “[00] [49] [00] = [00] [49] [00]” is established, only true is obtained.

  Note that, according to the COBOL grammar specification, when values are compared, 0 is supplemented in places where digits are insufficient. The end of both the left side and the right side is a sign part. For grammar specifications, see, for example, “NetCOBOL V10.5 COBOL Grammar Book (http://software.fujitsu.com/jp/manual/manualfiles/m120024/b1wd1361/06z000/b1wd-1361-06z0.pdf)” about.

  As described above, when the present embodiment described below is not used, in a situation where symbol values and specific values are mixed for the same variable (that is, symbol values and specific values exist in different digits), the transcription is appropriate. Will not be done. Furthermore, the comparison is not properly performed. That is, since it is determined that only true is established at a branch despite symbolic execution, it is not possible to cover the processing paths when it is false. As a result, test data cannot be extracted for the processing path in the case of false.

  Therefore, in the present embodiment, by holding variable values in the following format, it is possible to eliminate calculation errors in symbolic execution.

  FIG. 3 shows a block diagram of the information processing apparatus 10 in the present embodiment. The information processing apparatus 10 includes a program storage unit 6, a program reading unit 51, an instruction execution unit 52, a symbolic execution unit 5 including a counter control unit 53 and a recording unit 54, an execution history storage unit 4, an external register 3, It has a main storage device 2 and a symbol value calculation device 1.

  The program storage unit 6 stores a program to be subjected to symbolic execution. The program reading unit 51 reads the program stored in the program storage unit 6 and outputs it to the instruction execution device 52. The instruction execution device 52 executes a program and outputs a predetermined command (in this embodiment, a transfer command and a comparison command) to the symbol value calculation device 1. Further, the instruction execution device 52 stores the result of the operation in the program in the external register 3. When the execution of the program received from the program reading unit 51 is completed, the instruction execution device 52 notifies the counter control unit 53 to that effect. The counter control unit 53 manages a program counter and controls reading by the program reading unit 51. The recording unit 54 generates an execution history (for example, a path condition) based on the information received from the instruction execution device 52 and stores the execution history in the execution history storage unit 4. The symbol value calculation device 1 performs a transfer calculation using the data stored in the external register 3 and the data stored in the main storage device 2 and stores the result in the main storage device 2. In addition, the symbol value calculation device 1 performs a comparison calculation using data stored in the main storage device 2 and stores the result in the external register 3.

  FIG. 4 shows a block diagram of the symbol value computing device 1. The symbol value arithmetic apparatus 1 includes an instruction decoder 100, a reading unit 101, a transfer processing unit 110 including a first transfer processing unit 111 and a second transfer processing unit 112, a first format conversion unit 121, and a second format conversion unit. 122, a first conversion unit 120 including a third format conversion unit 131 and a fourth format conversion unit 132, a conversion unit 102, a data register 103, a writing unit 104, and a comparison operation unit. 105 and a mixed value generation unit 106.

  The instruction decoder 100 decodes the instruction received from the symbolic execution device 5 and specifies information of variables (in this embodiment, a transfer source variable and a transfer destination variable, or a comparison source variable and a comparison destination variable). Then, the instruction decoder 100 instructs the reading unit 101 to execute processing based on the variable information and the decoded instruction.

  The reading unit 101 reads the value of a variable from the external register 3 or reads the value of a variable from the main storage device 2 in accordance with an instruction from the instruction decoder 100. When the command is a transfer command, the reading unit 101 outputs the read value of the transfer source variable to the transfer processing unit 110. When the instruction is a comparison instruction, the reading unit 101 outputs the read values of the comparison source variable and the comparison destination variable to the comparison operation unit 105.

  The first transfer processing unit 111 outputs the value (specific value here) of the transcription source variable received from the reading unit 101 to the first format conversion unit 121 and the third format conversion unit 131. The second transfer processing unit 112 outputs the value of the transcription source variable (here, the symbol value) received from the reading unit to the second format conversion unit 122 and the fourth format conversion unit 132.

  The first format conversion unit 121 performs a process of converting the format of the specific value of the transfer source variable received from the first transfer processing unit 111 into a pack format (the pack format is the same as the internal decimal format). The result is output to the conversion device 102. The second format conversion unit 122 performs a process of converting the symbol value format of the transcription source variable received from the second transfer processing unit 112 into a digit number format, and outputs the processing result to the conversion device 102. The digit number format is a format having a value for each digit (that is, an array in which each element represents a single digit value).

  The third format conversion unit 131 uses the specific value of the transfer source variable received from the first transfer processing unit 111 and the information of the transfer destination variable received from the instruction decoder 100 to store the pack format stored in the data register 103. Process to convert the specific value into the format of the destination variable. Then, the third format conversion unit 131 outputs the processing result to the conversion device 102. The fourth format conversion unit 132 uses the symbol value of the transfer source variable received from the second transfer processing unit 112 and the information of the transfer destination variable received from the instruction decoder 100 to store the number of digits stored in the data register 103. The process of converting the symbol value of the transfer source variable to the format of the transfer destination variable. Then, the fourth format conversion unit 132 outputs the processing result to the conversion device 102.

  The conversion device 102 stores the processing result received from the first conversion unit 120 and the processing result received from the second conversion unit 130 in the data register 103.

  The writing unit 104 stores the data stored in the data register 103 in the main storage device 2 using the information of the transfer destination variable received from the instruction decoder 100.

  The comparison calculation unit 105 instructs the mixed value generation unit 106 to generate a mixed value. The mixed value generation unit 106 generates a mixed value for each of the comparison source variable and the comparison target variable and stores the mixed value in the data register 103. Further, the comparison operation unit 105 performs processing for comparing the mixed value of the comparison source variable and the mixed value of the comparison destination variable, and stores the processing result in the external register 3.

  Note that the thick arrow in FIG. 4 represents data reading, and the thick dotted arrow in FIG. 4 represents data writing.

  FIG. 5 shows an example of data stored in the main storage device 2. In the example of FIG. 5, for each variable, an address where a value is stored, a specific value corresponding to each address, and a symbol value corresponding to each address are stored. As described above, the main storage device 2 is provided with an area for storing a specific value and an area for storing a symbol value. Each of the addresses A1 to A8 is an address of a storage area having a capacity of 1 byte. S1 to S4, S6 to S8, and S10 to S12 represent symbol values, and F5, F9, and F13 represent values of the code part. In the internal binary format, a storage area is secured as two digits per byte. In FIG. 5, S11 stored at the address A7 corresponds to the 100th place of the variable C.

  FIG. 6 shows an example of data stored in the external register 3. In the example of FIG. 6, a specific value corresponding to each address and a symbol value corresponding to each address are stored. “NA” represents that the value is empty. Similar to the main storage device 2, the external register 3 is provided with an area for storing specific values and an area for storing symbol values.

  Next, the operation of the symbol value computing device 1 will be described with reference to FIGS.

  First, the instruction decoder 100 in the symbol value arithmetic unit 1 decodes the instruction received from the symbolic execution unit 5, and specifies information on the transfer source variable and the transfer destination variable, or information on the comparison source variable and the comparison destination variable ( FIG. 7: Step S1). The information specified in step S1 includes, for example, start address, data length, start position information, number of digits, decimal part, and variable format information.

  The reading unit 101 reads the value of the variable from the external register 3 or the main storage device 2 in accordance with the variable information received from the instruction decoder (step S3). Further, the reading unit 101 determines whether the instruction type is a transfer instruction (step S5). When it is a transfer command (step S5: Yes route), the reading unit 101 outputs the read value of the transfer source variable to the transfer processing unit 110. Further, the instruction decoder 100 outputs information on the transfer source variable and the transfer destination variable to the transfer processing unit 110, outputs information on the transfer source variable to the first conversion unit 120, and outputs information on the transfer destination variable to the second conversion unit. 130 and the writing unit 104.

  In response to this, the transfer processing unit 110 executes a concrete value transfer operation (step S7) and a symbol value transfer operation (step S9). First, a specific value transfer operation will be described with reference to FIG.

  First, the first transfer processing unit 111 in the transfer processing unit 110 outputs the specific value of the transcription source variable to the first format conversion unit 121 and the third format conversion unit 131. The first format conversion unit 121 converts the specific value of the transfer source variable into a pack format and stores it in the address RA of the data register 103 (FIG. 8: step S21). The storage in the data register 103 is performed by the conversion device 102.

  The first transfer processing unit 111 uses the information of the transfer destination variable received from the instruction decoder 100 to calculate the difference between the decimal value digits of the transfer source variable and the decimal value digits of the transfer destination variable, and Only the converted concrete value stored in the address RA is shifted (step S23). That is, a process for adjusting the position of the decimal point is executed.

  If no value is stored in the sign part of the transfer destination variable, the first transfer processing part 111 sets a positive sign value in the sign part of the specific value after the process of step S23 is executed ( Step S25).

  The first transfer processing unit 111 converts the specific value after the process of step S25 is executed into the number of digits that can be stored in the number of digits of the transfer destination variable (step S27).

  The third format conversion unit 131 converts the specific value after the processing of step S27 is performed into the format of the transfer destination variable using the information of the transfer destination variable received from the instruction decoder 100, and the address of the data register 103 The RA is overwritten (step S29). Then, the process returns to the caller process. Note that the conversion device 102 overwrites the data register 103.

  Next, the symbol value transfer operation will be described with reference to FIG.

  First, the second transfer processing unit 112 in the transfer processing unit 110 outputs the symbol value of the transcription source variable to the second format conversion unit 122 and the fourth format conversion unit 132. The second format conversion unit 122 converts the symbol value of the transfer source variable into a digit number format and stores it in the address RB of the data register 103 (FIG. 9: Step S31). The storage in the data register 103 is performed by the conversion device 102.

  Here, the conversion to the digit number format performed by the second format conversion unit 122 will be described with reference to FIG. First, the second format converter 122 calculates the number of addresses used for storage (FIG. 10: step S41). Here, the number of addresses used for storage is obtained by (((original data length) -1) / 2 + ((original data length) -1)% 2) +1. “%” Is an operator for obtaining a remainder, and a remainder when dividing by 2 by “% 2” can be obtained. Further, the second format converter 122 stores the converted symbol values in order from the last address in the address RB of the data register 103 (step S43). Then, the process ends.

  Returning to the description of FIG. 9, the second transfer processing unit 112 uses the information of the transfer destination variable received from the instruction decoder 100 to make a difference between the number of decimal digits of the transfer source variable and the number of decimal values of the transfer destination variable. And the converted symbol value stored in the address RB is shifted by the difference (step S33). That is, a process for adjusting the position of the decimal point is executed.

  If no value is stored in the sign part of the transfer destination variable, the second transfer processing part 112 sets a positive sign value in the sign part of the symbol value after the process of step S33 is executed ( Step S35).

  The second transfer processing unit 112 converts the symbol value after the process of step S35 is executed into the number of digits that can be stored in the number of digits of the transfer destination variable (step S37).

  The fourth format conversion unit 132 converts the symbol value after the processing of step S37 into the format of the transfer destination variable using the information of the transfer destination variable received from the instruction decoder 100, and the address of the data register 103 The RB is overwritten (step S39). Then, the process returns to the calling process. Note that the conversion device 102 overwrites the data register 103.

  Returning to the description of FIG. 7, the writing unit 104 writes the specific value and the symbol value stored in the data register 103 to the main memory 2 (step S11). Then, the process ends.

  As described above, the specific value and the symbol value are managed in a format having a value for each digit, and the two values are stored in different areas so that the specific value and the symbol value exist in different digits in the same variable. Even in such a case, an appropriate calculation can be performed.

  In addition, since any variable format is once converted into a common format, it is easy to share components used for computation.

  Further, in the example described above, when a specific value is transferred to a certain digit in the value of the variable, the symbol value is emptied. This makes it possible to determine whether each digit in the value of the variable is a specific value or a symbol value depending on whether the symbol value is empty.

  Here, a specific example is shown about the transcription | transfer operation demonstrated above.

For example, assume that the data shown in FIG. 5 is stored in the main storage device 2 and the data shown in FIG. 6 is stored in the external register 3. Suppose that there is the following transfer instruction.
-The specific value of the transfer source is that the start address is R1, the data length is 1, the start position is 1 (digit), the number of digits is 1, the decimal part is 0, and the format is external Decimal format.
The transfer destination variable (variable A) has a start address of A1, a data length of 4, a start position of 2, a number of digits of 1, a decimal part of 0, and a format of external decimal Format.

  FIG. 11 shows changes in specific values stored in the data register 103 in the situation described above. In the example of FIG. 11, the specific value [4] is stored in the data register 103 at the initial stage, and does not change after the processing of the first transfer processing unit 111 and the like thereafter.

  FIG. 12 shows changes in symbol values stored in the data register 103 in the situation described above. In the example of FIG. 12, the symbol value [NA] is stored in the data register 103 at the initial stage, and does not change after the processing of the second transfer processing unit 112 and the like thereafter. This is because the above-described instruction is a concrete value transfer instruction and the transfer source has no symbol value.

  FIG. 13 shows an example of data stored in the main storage device 2 when the transfer of the specific value in the transfer command described above is completed. In FIG. 13, the color different from the data shown in FIG. The specific value [4] in the data register 103 is stored at the address A2 in the main storage device 2. At this stage, no symbol value is stored.

  FIG. 14 shows an example of data stored in the main memory 2 when the symbol value transfer in the transfer instruction described above is completed. In FIG. 14, a color is given to a different part from the data shown in FIG. The symbol value [NA] in the data register 103 is stored at the address A2 in the main storage device 2. However, the specific value and the symbol value are stored in separate areas.

Further, for example, in the state where the data shown in FIG.
The specific value of the transfer source variable (variable A) is that the start address is A1, the data length is 4, the start position is 1, the number of digits is 4, the decimal part is 2, and the format is External decimal format.
-The transfer destination variable (variable B) has a start address of A5, a data length of 2, a start position of 1, a number of digits of 3, a decimal part of 2, and a format of internal decimal Format.

  FIG. 16 shows changes in specific values stored in the data register 103 in the situation described above. In the example of FIG. 16, the specific values [0] [4] [9] [00] are read from the main storage device 2 in the first stage, and [00] [49] are processed by the first format conversion unit 121. [00] is converted into a pack format. Further, it is converted into the format [49] [00] by the processing of the first transfer processing unit 111.

  FIG. 17 shows changes in symbol values stored in the data register 103 in the situation described above. In the example of FIG. 17, the symbol value [S1] [NA] [NA] [S4_F5] is read from the main storage device 2 at the first stage, and [NA_S1] [NA_NA] is processed by the second format conversion unit 122. [S4_F5] is converted into a digit number format. Further, it is converted into the format [NA_NA] [S4_F5] by the processing of the second transfer processing unit 112.

  FIG. 18 shows an example of data stored in the main storage device 2 when the transfer of the specific value in the transfer command described above is completed. In FIG. 18, the color different from the data shown in FIG. 15 is given. Since the variable B is in the internal decimal format, the specific value “4” and the specific value “9” are stored in the address A5. At this stage, no symbol value is stored.

  FIG. 19 shows an example of data stored in the main storage device 2 when the symbol value transfer in the transfer command described above is completed. In FIG. 19, colors are given to portions different from the data shown in FIG. 15. The symbol value “NA” corresponding to the specific value “4” and the symbol value “NA” corresponding to the specific value “9” are stored in the area for storing the symbol value at the address A5.

Further, for example, in the state where the data shown in FIG.
The specific value of the transfer source variable (variable A) is that the start address is A1, the data length is 4, the start position is 1, the number of digits is 4, the decimal part is 2, and the format is External decimal format.
-The transfer destination variable (variable C) has a start address of A7, a data length of 2, a start position of 1, a number of digits of 3, a decimal part of 1, and a format of internal binary Format.

  FIG. 20 shows changes in specific values stored in the data register 103 in the situation described above. In the example of FIG. 20, the specific values [0] [4] [9] [00] are read from the main storage device 2 in the first stage, and [00] [49] are processed by the first format conversion unit 121. [00] is converted into a pack format. Also, the first transfer processing unit 111 shifts right to form [04] [90] in order to align the decimal point. Furthermore, it is converted into the format of the transfer destination variable [0x00] [0x31] by the processing of the third format conversion unit 131.

  FIG. 21 shows changes in symbol values stored in the data register 103 in the situation as described above. In the example of FIG. 21, the symbol value [S1] [NA] [NA] [S4_F5] is read from the main storage device 2 in the first stage, and [NA_S1] [NA_NA] is processed by the second format conversion unit 122. [S4_F5] is converted into a digit number format. Further, the second transfer processing unit 112 shifts right to form [S1_NA] [NA_F5] in order to align the decimal point.

  FIG. 22 shows an example of data stored in the main storage device 2 when the transfer of the specific value in the transfer command described above is completed. In FIG. 22, the color different from the data shown in FIG. 19 is colored. Since the variable C is in the internal binary format, the specific value “0x00” is stored in the address A7, and the specific value “0x31” is stored in the address A8. At this stage, no symbol value is stored.

  FIG. 23 shows an example of data stored in the main storage device 2 when the symbol value transfer in the transfer instruction described above is completed. In FIG. 23, colors are given to portions different from the data shown in FIG. The symbol value “S1” and “NA” corresponding to the specific value “0x00” are stored in the area for storing the symbol value at the address A7, and the specific value “0x31” is stored in the area for storing the symbol value at the address A8. The symbol value “NA” corresponding to is stored.

  Returning to the description of FIG. 7, if the instruction type is not transcription in step S5 (step S5: No route), the instruction type is comparison. Therefore, the reading unit 101 outputs the read comparison source variable value and the comparison target variable value to the comparison operation unit 105. Then, the comparison calculation unit 105 executes comparison calculation (step S13). The comparison operation will be described with reference to FIGS. 24 and 25.

  First, the comparison operation unit 105 outputs the value of the comparison source variable and the value of the comparison destination variable to the mixed value generation unit 106. In response, the mixed value generation unit 106 executes a mixed value generation process for each of the comparison source variable and the comparison target variable (FIG. 24: step S51). The mixed value generation process will be described with reference to FIG.

  First, the mixed value generation unit 106 converts the format of the symbol value of the comparison source variable and the symbol value of the comparison destination variable into a digit number format, and stores it in the address RA of the data register 103 (FIG. 25: step S61). In step S <b> 61, the mixed value generation unit 106 causes the second format conversion unit 122 in the first conversion unit 120 to perform conversion.

  The mixed value generation unit 106 converts the format of the specific value of the comparison source variable and the specific value of the comparison destination variable into a pack format, and stores it in the address RB of the data register 103 (step S63). In step S63, the mixed value generation unit 106 causes the first format conversion unit 121 in the first conversion unit 120 to perform conversion.

  The mixed value generation unit 106 determines whether there is an empty digit in the symbol value of the comparison source variable and the symbol value of the comparison target variable stored in the address RA (step S65). If there is no empty digit (step S65: No route), the process returns to the caller process. On the other hand, when there is an empty digit (step S65: Yes route), the same digit value in the specific value stored in the address RB is set to the empty digit (step S67). Then, the process returns to the calling process. For example, when the third digit in the symbol value of the comparison source variable is empty, the third digit value in the specific value of the comparison source variable is set. For example, when the second digit in the symbol value of the comparison target variable is empty, the second digit value in the specific value of the comparison target variable is set. As described above, a concrete value exists in an empty digit among the digits of the symbol value.

  By executing the processing as described above, it is possible to generate values in which specific values and symbol values exist in different digits.

  Returning to the description of FIG. 24, the comparison operation unit 105 calculates the difference in the number of decimal digits for the two mixed values generated by the mixed value generation process, and the smaller the number of decimals by the difference. The mixed value is shifted to the left (step S53). That is, processing for adjusting the position of the decimal point is executed.

  When there is no sign in either the comparison source mixed value or the comparison target mixed value (that is, the sign part has no value), the comparison operation unit 105 sets a positive sign value in both sign parts. (Step S55).

  The comparison calculation unit 105 determines a satisfiability possibility (step S57). Note that the satisfiability may be determined using a constraint solver or the like. For the constraint solver, see, for example, Non-Patent Document 1 shown above.

  The comparison calculation unit 105 stores the determination result of satisfiability in the external register 3 (step S59). And it returns to description of FIG. 7 and complete | finishes a process.

  FIG. 26 shows an example of data stored in the external register 3 by the process of step S59. In the example of FIG. 26, information indicating whether or not the state where the branch condition is true is stored in the address RT, and information indicating whether or not the state where the branch condition is false is satisfied in the address RF. Is stored. In the example of FIG. 26, “1” indicates that the condition is satisfied, and “0” indicates that the condition is not satisfied. Therefore, in this example, both the state where the branch condition is true and the state where it is false are satisfied.

  For example, the data shown in FIG. 23 is stored in the main memory 2 by the transfer operation of “MOVE variable A TO variable B” on line 140 and “MOVE variable A TO variable C” on line 150 of the program shown in FIG. Suppose that Then, using the data, the comparison between the mixed value [0_S1] [4_9] [S4_F5] and the mixed value [4_9] [S4_F5] is performed in the “IF (variable A = variable B) THEN” on the 160th line. To do. Here, if the value of S1 is other than 0 (variable A = variable B), the branch condition is false, and if the value of S1 is 0 (variable A = variable B), the branch condition is true.

  Also, in line 190 “IF (variable A = variable C) THEN”, the mixed value [0_S1] [4_9] [S4_F5] ”and the mixed value [0_S1] [4_9] [0_F5]” are compared. . Here, if the value of S4 is other than 0 (variable A = variable C), the branch condition is false, and if the value of S4 is 0 (variable A = variable C), the branch condition is true.

  Therefore, according to the method of the present embodiment, it is determined that both true and false are satisfied for the branch condition, and both processing paths can be executed. As a result, the processing paths can be covered in symbolic execution, so that there is no omission of test data extraction.

  Here, a specific example of the mixed value generation described above will be shown.

  FIG. 27 shows an example of mixed value generation for variable A. In the example of FIG. 27, it is assumed that the data shown in FIG. In this case, first, the specific value [0] [4] [9] [00] of the variable A is read from the main storage device 2, and the symbol value [S1] [NA] [00] of the variable A is read from the main storage device 2. NA] [S4_F5] is read. The specific values [0] [4] [9] [00] are converted into a pack format [00] [49] [00] by the processing of the first format conversion unit 121. Further, the symbol values [S1] [NA] [NA] [S4_F5] are converted into a digit number format of [NA_S1] [NA_NA] [S4_F5] by the processing of the second format converter 122. Then, when the mixed value generation unit 106 generates a mixed value based on the specific values [00] [49] [00] and the symbol values [NA_S1] [NA_NA] [S4_F5], [0_S1] [4_9] [S4_F5] is obtained. .

  FIG. 28 shows an example of mixed value generation for variable C. In the example of FIG. 28, it is assumed that the data shown in FIG. In this case, first, the specific value [0x00] [0x31] of the variable C is read from the main storage device 2, and the symbol value [S1] [NA] [NA] [S4_F5] of the variable C is read from the main storage device 2. Read out. The specific value [0x00] [0x31] is converted into a pack format [04] [90] by the processing of the first format conversion unit 121. Further, the symbol value [S1] [NA] [NA] [S4_F5] is converted into a digit number format of [S1_NA] [NA_F5] by the processing of the second format conversion unit 122. Then, when the mixed value generation unit 106 generates a mixed value based on the specific value [04] [90] and the symbol value [S1_NA] [NA_F5], [S1_4] [9_F5] is obtained. When the comparison operation unit 105 executes the process of matching the mixed value for the variable A and the position of the decimal point, the mixed value finally has the format [0_S1] [4_9] [0_F5].

  Although one embodiment of the present invention has been described above, the present invention is not limited to this. For example, the functional block configuration of the information processing apparatus 10 described above may not match the actual module configuration.

  Further, the configuration of each table described above is an example, and the configuration as described above is not necessarily required. Further, in the processing flow, the processing order can be changed if the processing result does not change. Further, it may be executed in parallel.

  For example, in the example described above, the symbol value transfer process is executed after the specific value transfer process is executed, but both may be executed in parallel.

  In addition, although the format for holding a 2-digit value in one byte is the digit number format, the digit number format may be changed depending on the format supported by the language. For example, a format in which a 1-digit value is stored in 1 byte or a format in which a 4-digit value is stored in 1 byte may be used.

  The embodiment of the present invention described above is summarized as follows.

  The arithmetic device according to the first aspect of the present embodiment converts (A) the specific value of the first variable that is the target of the calculation in symbolic execution into the first format that has a value for each digit. In addition, a first conversion unit that converts the symbol value of the first variable into a second format that has a value for each digit, and (B) a specific value after conversion by the first conversion unit, And a first storage processing unit that stores the symbol value after conversion by the first conversion unit in a second storage region different from the first storage region.

  In this way, it is possible to appropriately perform variable operations in symbolic execution. More specifically, even when the specific value and the symbol value exist in different digits in the same variable, it is possible to eliminate an operation error. Furthermore, since the variable is converted into a common format no matter what the format is, it is easy to share the parts used for the transfer operation or the like.

  In addition, the arithmetic device described above converts (C) the specific value stored in the first storage area into the format of the second variable that is the transfer destination variable, and stores it in the second storage area. A second conversion unit that converts the converted symbol value into a second variable format; and (D) a specific value after conversion by the second conversion unit is stored in the third storage area, and by the second conversion unit. You may further have the 2nd storage process part which stores the symbol value after conversion in the 4th storage area different from a 3rd storage area. In this way, a transcription operation can be performed between variables in symbolic execution.

  In addition, the arithmetic unit described above (E) performs processing for adjusting the position of the decimal point to the second variable for each of the concrete value and the symbol value before the conversion by the second conversion unit, and the number of digits. You may further have a 1st process part which performs the process which adjusts to the number of digits of a 2nd variable. Thereby, a transcription | transfer operation comes to be performed appropriately.

  In addition, the arithmetic unit described above replaces (F) an empty digit among the digits of the symbol value stored in the second storage area with the same digit of the specific value stored in the first storage area. To generate the first mixed value, convert the specific value stored in the fifth storage area into the first format, and convert the symbol value stored in the sixth storage area into the second format. A generation unit that generates a second mixed value by replacing an empty digit in the converted symbol value digits with the same digit in the converted concrete value, and (G) the first mixed value and the first You may further have a comparison part which compares with 2 mixed values. In this way, comparison operations can be performed between variables in symbolic execution.

  Further, the comparison unit described above may perform (g1) a process of matching the decimal point position of the first mixed value with the decimal point position of the second mixed value. In this way, the comparison operation is appropriately performed.

  The calculation method according to the second aspect of the present embodiment converts (H) the specific value of the first variable that is the object of calculation in symbolic execution into the first format that has a value for each digit. (I) a symbol value of the first variable is converted to a second format that has a value for each digit, and is different from the first storage region. Including the process of storing in the storage area.

  A program for causing a computer to perform the processing according to the above method can be created. The program can be a computer-readable storage medium such as a flexible disk, a CD-ROM, a magneto-optical disk, a semiconductor memory, a hard disk, or the like. It is stored in a storage device. The intermediate processing result is temporarily stored in a storage device such as a main memory.

  The following supplementary notes are further disclosed with respect to the embodiments including the above examples.

(Appendix 1)
The specific value of the first variable that is the object of the operation in the symbolic execution is converted to the first format that has a value for each digit, and the symbol value of the first variable is changed to a value for each digit. A first conversion unit for converting to a second format which is a format having
A specific value after conversion by the first conversion unit is stored in a first storage area, and a symbol value after conversion by the first conversion unit is stored in a second storage area different from the first storage area. A first storage processing unit for storing in
A computing device having

(Appendix 2)
The specific value stored in the first storage area is converted into the format of a second variable that is a transfer destination variable, and the symbol value stored in the second storage area is converted into the second variable. A second conversion unit for converting to the format of
The specific value after conversion by the second conversion unit is stored in a third storage area, and the symbol value after conversion by the second conversion unit is stored in a fourth storage area different from the third storage area. A second storage processing unit for storing;
The arithmetic unit according to appendix 1, further comprising:

(Appendix 3)
Before executing the conversion by the second conversion unit, for each of the specific value and the symbol value, a process for adjusting the position of the decimal point to the second variable and setting the number of digits to the number of digits of the second variable The arithmetic unit according to supplementary note 2, further comprising: a first processing unit that performs a matching process.

(Appendix 4)
Generating a first mixed value by replacing an empty digit among the digits of the symbol value stored in the second storage area with the same digit of the specific value stored in the first storage area; The specific value stored in the fifth storage area is converted into the first format, the symbol value stored in the sixth storage area is converted into the second format, and the digit of the symbol value after conversion A generation unit that generates a second mixed value by replacing an empty digit by the same digit in the converted concrete value;
A comparison unit for comparing the first mixed value and the second mixed value;
The arithmetic unit according to any one of appendices 1 to 3, further comprising:

(Appendix 5)
The comparison unit includes:
The arithmetic unit according to appendix 4, wherein a process of matching a decimal point position of the first mixed value with a decimal point position of the second mixed value is performed.

(Appendix 6)
The specific value of the first variable that is the target of the operation in symbolic execution is converted into a first format that has a value for each digit, and is stored in the first storage area.
The arithmetic unit executes a process of converting the symbol value of the first variable into a second format which is a format having a value for each digit and storing the symbol value in a second storage area different from the first storage area Calculation method to be performed.

DESCRIPTION OF SYMBOLS 1 Symbol value arithmetic unit 2 Main memory unit 3 External register 4 Execution history storage part 5 Symbolic execution unit 51 Program reading part 52 Instruction execution unit 53 Counter control part 54 Recording part 6 Program storage part 10 Information processing apparatus 100 Instruction decoder 101 Reading part 102 conversion device 103 data register 104 writing unit 105 comparison operation unit 106 mixed value generation unit 110 transfer processing unit 111 first transfer processing unit 112 second transfer processing unit 120 first conversion unit 121 first format conversion unit 122 second format conversion Unit 130 second conversion unit 131 third format conversion unit 132 fourth format conversion unit

Claims (4)

The specific value of the first variable that is the target of the operation in symbolic execution is converted to the first format that has a value for each digit, and the symbol value of the first variable has a value for each digit. Is converted to the second format, and the specific value of the second variable that is the target of the operation in the symbolic execution is converted to the first format, and the symbol value of the second variable is converted to the second variable. A first conversion unit for converting to a format of 2 ,
The first specific value, which is the specific value after conversion of the specific value of the first variable , is stored in the first storage area, and the first symbol value after conversion of the symbol value of the first variable is the first value A symbol value is stored in a second storage area different from the first storage area, and a second specific value that is a converted specific value of the second variable is stored in a third storage area. A first storage processing unit for storing a second symbol value, which is a symbol value after conversion of the symbol value of the second variable, in a fourth storage area different from the third storage area ; ,
An empty digit among the digits of the first symbol value stored in the second storage area is replaced with the same digit of the first concrete value stored in the first storage area. 1 is generated, and an empty digit among the digits of the second symbol value stored in the fourth storage area is replaced with the second specific value stored in the third storage area. Generating a second mixed value by replacing with the same digit of
A comparison unit for comparing the first mixed value and the second mixed value;
A computing device having Wherein the first of said first specific value stored in the storage area, and converts the third type of variable is a transfer destination variable, the second of said first symbol stored in the storage area A second conversion unit for converting a value into the format of the third variable;
The first specific value after conversion by the second conversion unit is stored in a fifth storage area, and the first symbol value after conversion by the second conversion unit is stored in the fifth storage area. A second storage processing unit for storing in a different sixth storage area;
The arithmetic device according to claim 1, further comprising: The comparison unit is
A process of matching the decimal point position of the first mixed value with the decimal point position of the second mixed value.
The arithmetic device according to claim 1 or 2. The specific value of the first variable that is the target of the operation in the symbolic execution is converted to the first format that has a value for each digit, and the first specific value that is the converted specific value is the first value Store it in the storage area,
The symbol value of the first variable is converted to a second format that has a value for each digit, and the first symbol value that is the converted symbol value is different from that of the first storage area. Stored in the second storage area ,
An empty digit among the digits of the first symbol value stored in the second storage area is replaced with the same digit of the first concrete value stored in the first storage area. Produces a blend value of 1
Converting the specific value of the second variable that is the object of the calculation in the symbolic execution into the first format, storing the second specific value that is the converted specific value in the third storage area,
The symbol value of the second variable is converted into the second format, and the converted second symbol value is stored in a fourth storage area different from the third storage area. ,
An empty digit among the digits of the second symbol value stored in the fourth storage area is replaced with the same digit of the second concrete value stored in the third storage area. Produces a mixed value of 2
Making a comparison between the first blend value and the second blend value;
An arithmetic method in which the arithmetic device executes processing.

Images